Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Bioinspired Implementation and Assessment of a Remote-Controlled Robot

Bioinspired Implementation and Assessment of a Remote-Controlled Robot Hindawi Applied Bionics and Biomechanics Volume 2019, Article ID 8575607, 10 pages https://doi.org/10.1155/2019/8575607 Research Article Bioinspired Implementation and Assessment of a Remote- Controlled Robot 1,2 3 Yves Rybarczyk and Diogo Gil Carvalho Intelligent & Interactive Systems Lab (SI2 Lab), Universidad de Las Américas, 170125 Quito, Ecuador Dalarna University, 791-88 Falun, Sweden Department of Electrical Engineering, CTS/UNINOVA, Nova University of Lisbon, 2829-516 Monte de Caparica, Portugal Correspondence should be addressed to Yves Rybarczyk; y.rybarczyk@fct.unl.pt Received 26 February 2019; Revised 9 May 2019; Accepted 21 August 2019; Published 11 September 2019 Guest Editor: Francesca Cordella Copyright © 2019 Yves Rybarczyk and Diogo Gil Carvalho. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Daily activities are characterized by an increasing interaction with smart machines that present a certain level of autonomy. However, the intelligence of such electronic devices is not always transparent for the end user. This study is aimed at assessing the quality of the remote control of a mobile robot whether the artefact exhibits a human-like behavior or not. The bioinspired behavior implemented in the robot is the well-described two-thirds power law. The performance of participants who teleoperate the semiautonomous vehicle implementing the biological law is compared to a manual and nonbiological mode of control. The results show that the time required to complete the path and the number of collisions with obstacles are significantly lower in the biological condition than in the two other conditions. Also, the highest percentage of occurrences of curvilinear or smooth trajectories are obtained when the steering is assisted by an integration of the power law in the robot’s way of working. This advanced analysis of the performance based on the naturalness of the movement kinematics provides a refined evaluation of the quality of the Human-Machine Interaction (HMI). This finding is consistent with the hypothesis of a relationship between the power law and jerk minimization. In addition, the outcome of this study supports the theory of a CNS origin of the power law. The discussion addresses the implications of the anthropocentric approach to enhance the HMI. 1. Introduction the robot seems relevant. For instance, the human-robot interaction tends to be improved when the machine has a Industries face an increasing demand for collaborative robots humanoid appearance [6]. This fact can be explained by an inconscient tendency of the human being to anthropomor- that exhibit human-like behaviors. This trend is justified by the fact that it is easier for an operator to predict the actions phize the artefacts they interact with, in order to predict their of a robot that behaves more like a human being than like a behavior and increase their acceptance of the machine [7]. machine [1]. A study that uses the experimental paradigm Besides the situation of interaction, the implementation of Motor Interference (MI) shows that the motor perfor- of human-like behaviors in a robot’s way of working also mance of an individual can be influenced by the perception seems to benefit an operator that has to control a machine. of the movements of a robot, if the machine replicates some This statement is particularly true in the context of teleopera- characteristics of biological motion [2–4]. In particular, it tion, which implies several limitations for a human operator. seems that the movement velocity profile is sufficient to cre- For instance, the sensorial information received by the teleo- ate this interference. This result suggests that a movement perator can be altered, for example, the field of view is can be processed as biologic by the human brain, even if it reduced, not all the sensorial modalities are restituted (e.g., is not produced by a living being, on the condition that the audition and proprioception), and the response of the system artefact motion simulates (even approximatively) certain is delayed. Another aspect is the necessity to build or accom- biological kinematics [5]. Moreover, the physical aspect of modate new motor schemes to be able to control the user 2 Applied Bionics and Biomechanics aimed at assessing the motor control of a robot arm to interface of the device, which augments the mental workload. A promising approach to reduce the gap between the user assist surgeons [15]. The results show that both smooth- and the telerobot is to implement human-like behaviors in ness and minimum jerk are significant measures of exper- a robot [8, 9]. For instance, Rybarczyk et al. [10, 11] have tise levels. The end-effector trajectory evolves from sharp studied the effect of the implementation in a mobile robot and jerky in novices to smooth in experts. Thus, the of the human behavior of visuomotor anticipation over the authors conclude that these two features are excellent cri- locomotion, in which the direction of the robot pan-tilt teria to evaluate motor skill in the conditions of human- camera is automatically oriented toward the tangent point robot interaction. Although the power law is also identi- of the inside curve of the path, as walkers/cyclists/drivers fied as a discriminant measure of expertise, registering do [12–14]. The results show that the motor performances such a biological law seems to depend on the characteris- tics of the artefact. For example, some studies have dem- of the teleoperators are enhanced when they steer the bioinspired robot. A correlation between the replication of onstrated that this law is replicated in situations of biological laws and the level of expertise is also observed in teleoperation [11] and use of prostheses [28]. the case of the telemanipulation of robotic arms, such as in Actually, there is a controversy regarding the origins telesurgery [15]. and the violations of the 2/3 power law during the execu- Different strategies are used to implement human-like tion of the biological movements [22, 29–31]. On the one behaviors in a robot. A traditional approach applied in the hand, some studies tend to demonstrate that the power industry is to create anthropomorphic collaborative robots law is a signature of the Central Nervous System (CNS) (or cobots) that are trained to imitate biological motions, [32–34], because it seems to be independent of the dynam- through machine learning algorithms [16]. In the case of ics of the limbs. This law is indeed observed in a wide vari- the teleoperation, it seems that individuals feel also more ety of activities such as drawing [21], walking [35], and comfortable to control an anthropomorphic robot arm in smooth pursuit eye [36]. On the other hand, different stud- which the motion trajectory of the end effector is like a bio- ies defend a biomechanical [30] or, even, an artefactual logical movement [17]. Jerk minimization is one of the prin- explanation [37, 38]. There are also contradictory results cipal human-like behaviors that has been implemented to regarding the relationship between smoothness, minimum model a natural trajectory planning [18, 19]. The minimum jerk, and power law. Some studies show evidences that jerk is characterized by a bell-shaped velocity profile, in these features are related to each other [32, 39], whereas which the movement speed increases progressively, reaches others suggest the contrary [15, 29]. a peak near the midpoint, and then deceases slowly. This The present work attempts to tackle these different absence of abrupt changes seems to support the execution contradictory findings about the 2/3 power law by inte- of a smooth motion [20]. Another fundamental motor grating this bioinspired kinematics in a remote-controlled behavior is the relationship between the velocity and the mobile robot. An experiment is designed to compare the curvature of the biological movements, which is known teleoperation of a robot with the 2/3 power law (biological as the two-thirds power law [21, 22]. This law states that condition) versus two modes of control that do not imple- the angular velocity of the end effector is proportional to ment this human-like behavior (manual condition and the two-third root of its curvature or, equivalently, that nonbiological condition). In the biological mode, the the instantaneous tangential velocity (v ) is proportional engine speed is automatically servo controlled by the vehi- to the third root of the radius of curvature (r ), as cle trajectory according to the power law equation. In the manual mode, the user has to control both the velocity described in equation (1). In other words, it means that the velocity of the movement decreases in the highly and the direction of the mobile device. In the third condi- curved parts of the trajectory and increases when the tra- tion, the vehicle speed is also automatic, but the calcula- jectory becomes straighter. Implementing this model in a tion of the relationship between geometry and kinematics mobile robot tends to improve the raw performance when violates the biological motion. This last condition is used as a control to make sure that the potential difference of steering the vehicle [23]. performance between the two main conditions (biologic −1/3 vs. manual) is not caused by a dissimilar complexity of v = kr 1 t t the task (i.e., number of parameters that must be con- Nevertheless, few studies are interested in considering trolled by the participants). We posit the hypothesis that semiautonomous driving, in which the velocity is automat- refined features to gauge the quality of the Human- Machine Interaction (HMI). Instead of focusing only on ically set according to the power law principles (biological the raw performance (e.g., completion time of the task mode), should promote a significantly faster, safer, and and percentage occurrence of errors), these studies analyze more natural steering than the nonassisted (manual mode) the kinematics of the robot control [24–26]. To proceed and nonbiologic (artificial mode) control. The quality of the interaction is assessed from both the raw performance with such an advanced assessment, the human behaviors are now used as criteria to estimate an appropriate inter- (completion time and number of collisions) and refined parameters based on the smoothness of the trajectories. action. For instance, minimum jerk, smoothness, and 2/3 power law can be applied as a reference to evaluate a suit- The remainder of the manuscript is organized into able interaction between a human operator and an artefact three main sections. First, the implementation of the tele- operation system is described. The experimental protocol [17, 27]. These three features are compared in a study that Applied Bionics and Biomechanics 3 Motor Motor Motor (a) (b) Figure 1: (a) Schematic representation of the robot (top view) designed for the study. Two independent motors drive the front wheels, and a third one controls the rotation of the pan camera. This mobile vision is implemented by default to promote a visual anticipation over the change of direction. (b) Illustration of the camera behavior in some specific locations of the path. The blue arrow indicates the instantaneous direction of the vehicle and the red arrow represents the orientation of the pan camera at the same moment. It is notable that the angle between the two arrows is inversely proportional to the radius of curvature of the robot trajectory. The more curved is the shape of the path (e.g., position 2), the larger is the angle between the orientation of the camera and the direction of the vehicle, and vice versa (e.g., position 4). 2.2. Robot Behavior. The vehicle is built on four wheels, and conditions (manual vs. nonbiologic vs. biologic) are also explained in detail. Second, the results of the perfor- employing a front-wheel-drive system (Figure 1(a)). The mance for each condition are presented, analyzed, and dis- two front wheels are moved by two independent motors. cussed. Finally, the outcomes are interpreted, in order to The differential of speed between the right and the left draw some conclusions and perspectives regarding the appli- wheel rotation allows the vehicle to turn. The pan camera is set on a mobile structure, which is moved by another cation of the anthropocentric approach in the human-robot interaction, as well as the origins of the power law and its motor. The orientation of the camera is determined auto- relationship with jerk minimization. matically based on the direction of the robot, that is, the camera points toward the inside of the vehicle trajectory. Since any change of direction is systematically anticipated 2. Material and Methods by a rotation of the camera proportional to the curvature of the vehicle trajectory, a visual prediction over the robot 2.1. System Architecture. The three main elements that com- motion is provided to the operator. This mechanism pose the system are (i) a NXT mobile robot, (ii) an Android inspired from the human behavior [12, 14] is implemented device for the remote control, and a pan IP camera. Since by default, because it facilitates the teleoperation [8, 10]. the experiment is carried out in a situation of teleopera- Figure 1(b) shows examples of this visuolocomotor cou- tion (i.e., indirect perception and action on the robot envi- pling between camera and robot for different curves of ronment), a wireless connection is used to support the the path. communication between the principal components of the architecture. Two different protocols of communication 3. Experimental Conditions are applied. The Android-based remote control communi- cates with the NXT through Bluetooth technology. In 3.1. Manual Condition. Both speed and direction of the addition, the connection between the IP camera and the vehicle are manually controlled by the operator in this smartphone is supported by Wi-Fi communication. The experimental condition. Concentric semicircles that corre- robot is connected to the IP camera thanks to a support spond to different speed levels are displayed on the control library that permits the system integration between the panel of the user’s interface (Figure 2(a)). The bigger is the two entities. Thus, the operators use the Android remote radius of the semicircle, the higher is the speed. Thus, the control device to interact with the whole system, which vehicle velocity is calculated based on the distance between allows them to steer the mobile robot and receive a visual the center of all concentric semicircles and the selected feedback from the pan IP camera. An Android application semicircle. The direction of the robot is determined by is developed and implemented on the smartphone to per- the angle between the vertical of the screen and the loca- mit such an interaction. The tactile user interface enables tion of the user’s fingertip. The range of angles goes from ° ° the operator to control the trajectories of the vehicle, to 0 to 180 , rotating counter clockwise. If the fingertip of ° ° choose the steering mode of the robot (manual vs. nonbio- the user is positioned between 0 and 90 , the robot turns logic vs. biologic), to calibrate the pan camera, and to turn right, with a curvature proportional to the angle between the system on or off. the vertical (90 ) and the position of the finger (the more 4 Applied Bionics and Biomechanics (a) (b) Figure 2: (a) Representation of the GUI for the manual mode of driving. Each concentric circle represents a different speed (the larger the radius of the semicircle, the higher the velocity). (b) User interface for the biological and nonbiological conditions. A single semicircle enables the user to directly control the direction of the robot and indirectly set the speed of the vehicle. the location of the finger tends to 0 , the more the vehicle angle should perfectly fit to the human’s skills. On the turns right). On the contrary, if the position of the finger contrary, because of its unnatural behavior, the nonbiolog- ° ° is between 90 and 180 , the vehicle turns left (again, the ical semiautonomous control should be more challenging radius of the curvature of the trajectory depends on the for the teleoperator. Figure 3 summarizes, through a block diagram, the differences between the manual and semiau- angle from the vertical of the screen). The controller of the robot is constantly waiting for an input sent from tomatic modes of control. the graphic user interface, in order to update the direction and speed of the mobile platform. 4. Experimental Protocol 3.2. Biological and Nonbiological Conditions. In these driv- Thirty people (15 males and 15 females; 23 5± 3 5 years) ing modes, the user has only to use the touchscreen inter- took part in the experiment. All the participants had a nor- face to control the trajectory of the robot. The speed is mal or corrected-to-normal vision. The procedure con- automatically set according to the direction of the vehicle. formed to the Declaration of Helsinki and was approved In the biological condition, the 2/3 power law is used to by the Ethical Review Board of the Nova University of calculate the speed, which is based on the instantaneous Lisbon. The experiment was carried out in a classroom, radius of the curvature of the robot trajectory. The maxi- where the subjects had to teleoperate the NXT vehicle mum velocity of the robot is 30 cm/s, if the vehicle goes through the Android-based mobile device. The instructions straight forward. In the case that the radius of curvature provided to the participants were to steer the robot as safe decreases (to the right or to the left), the robot’s speed (the least collisions) and fast (the minimum completion diminishes by a rate of one-third (see equation (1)). In time) as possible through a path delimited by plastic the nonbiological condition, the velocity of the vehicle is blocks. The entire distance of the route was approximately also automatic, but it is not set according to the biological seven meters and consisted of numerous bends and motion. The relationship between speed and geometry changes in direction (curves and countercurves). The does not follow a power law, but a linear law described in sequence of the course was as follows: (i) a straight line, ° ° (ii) an approximately 150 bend, (iii) a 90 reverse curve, v = kr 2 t t (iv) another 150 bend, and (v) a final straight line (Figure 4). A blue adhesive strip marked the starting and Since it is not necessary to modulate manually the finishing line. The symmetric shape of the setup was espe- velocity, the graphic user interface is represented only by cially designed to carry out the route in both directions, a single semicircle (Figure 2(b)). The semicircle allows clockwise and anticlockwise. the operator to control the trajectory of the robot. From After a training session, all the subjects had to execute the user’s perspective, the way to steer the vehicle is iden- the trial twelve times: four repetitions in the manual tical to the manual mode of driving. The user has to inter- mode, four repetitions in the biological mode, and four act with the left and right portion of the semicircle to turn repetitions in the nonbiological mode. The order of the left and right, respectively. The more the fingertip is experimental conditions was counterbalanced from one located to the extremities of the semicircle, the more the subject to another so that ten individuals started with robot turns sharply. The only difference between these the manual control, ten others started with the biological two semiautomatic modes and the manual one is the fact control, and the last ten started with the nonbiological that the velocity is indirectly and automatically set when control. This counterbalancing was implemented to pre- the user chooses a determined direction. Precisely, the vent a possible learning effect, which would bias the out- robot speed is proportional to the selected steering angle. come of the study. For each of the principal conditions Thus, if the power law is adapted to the remote control (manual vs. biologic vs. nonbiologic), the trial was per- of an artefact, the matching between speed and steering formed twice clockwise and twice anticlockwise. The Applied Bionics and Biomechanics 5 Manual Semiautomatic Operator mode modes Visual n feedback Direction (휃) Linear Direction d speed (휈) (휃) Linear speed (휈) 휈 =f(휃 ) O t t Bluetooth Wi-Fi transmission protocol (IP camera) Microcontroller Motor Camera driver motor Left wheel Right wheel motor motor Vehicle motion Figure 3: Block diagram of the two modes of remote control. If the operator picks the manual mode (left side), the speed and direction are controlled independently. On the contrary, if a semiautomatic mode is selected (right side), the robot speed is automatically calculated from the power law (biological condition) or linear law (nonbiological condition) function of the direction defined by the user. completion time, the number of collisions, and the robot ishing line diminishes significantly from session 1 to ses- trajectory were recorded at the end of each trial. sion 4. No interaction effects are detected between the sessions (1, 2, 3, and 4) and the main conditions (manual, biologic, and nonbiologic). 5. Results In addition, the comparison of the completion time between the three conditions indicates a significant differ- The experimental data are statistically analyzed through ANOVA tests for multivariable comparisons and t-tests for ence (p < 005). As shown in Figure 5, the participants complete the task faster in the biological mode than in the pairwise comparisons. the manual (p < 01) and nonbiological (p < 005) steering 5.1. Completion Time. We first analyzed the time perfor- modes. The pairwise analyses confirm the significant dif- ference in session 1 (p < 01), session 2 (p < 05), and session mance of the participants to complete the task. Results indi- cate that the completion time is significantly affected by 3(p < 04). Nevertheless, this statistical difference vanishes in session 4, although the manual and nonbiological modes the experimental sessions (p < 05). A pairwise analysis shows a significant difference between session 1 and ses- tend to remain slower than the biological. The reduction of sion 4 (p < 03). This outcome indicates that the required the completion time over the sessions can be explained by a learning effect that occurs in all the conditions. time to guide the vehicle from the starting line to the fin- 6 Applied Bionics and Biomechanics 5.5 4.5 3.5 2.5 1.5 Sessions Figure 4: Picture of the experimental setting. The symmetric form Manual of the path was chosen to easily alternate the course direction of Biologic the robot from one trial to the next: once clockwise and once Nonbiologic counter clockwise. This alternation was designed to minimize the environment learning and a consequent machine-like driving ° Figure 6: Representation of the average number of collisions for of the vehicle. The two straight lines, two 150 bends, and one each of the main conditions (manual vs. biologic vs. nonbiologic) 90 bend are identified by broken yellow, green, and magenta against the four experimental sessions. lines, respectively. Note that these colors are added for a better understanding of the setup but were not visible during the experiment. ological modes to get steering skills as good as in the bio- logical condition. 5.3. Trajectory Smoothness. The last results address the ques- tion of the movement kinematics through the analysis of the jerk in the control of the robot trajectory. One way to quan- 42 tify the path smoothness is to calculate the instantaneous radius of curvature of each trajectory, then to evaluate the distribution frequency of the radius for all trials [40]. More specifically, the curve radius (r) is computed from the instan- taneous linear velocity (v) divided by the instantaneous rota- tion speed (w), as described in Sessions m/s r = 3 Manual radians/s Biologic Nonbiologic Subsequently, the radius of curvature is converted into a Figure 5: Representation of the mean completion time (in seconds) decimal logarithm. Therefore, if the vehicle has a low linear for each of the main conditions (manual vs. biologic vs. speed and a high velocity of rotation, the curve radius is very nonbiologic) against the four experimental sessions. small (<2), and gets smaller as the velocity of rotation increases. The result is a logarithmic value of r that is around 5.2. Number of Collisions. The assessment of the rate of colli- zero. Conversely, if the vehicle combines a translation and a sions was also performed to complement the analysis. The rotation (curvilinear trajectory), the curve radius is high statistical results indicate that the mean number of collisions (≥2) and its logarithm becomes superior to zero. A steering is significantly different over the sessions (p < 03). The pair- control in which the subject stops and turns in place provides wise analysis shows a significant diminution of the collisions a bimodal distribution of the curve radii, with one peak cen- from session 1 to session 4 (p < 02). These outcomes point tered on null values of the logarithm and another peak cen- out that the subjects have improved the quality of their driv- tered on positive values. On the contrary, a curvilinear (or ing skills over the experiment. There is no effect of interac- smooth) trajectory is characterized by a unimodal pattern tion between the four sessions and the main experimental of distribution centered on a value of the logarithm of the conditions (manual, biologic, and nonbiologic). radius of curvature higher than zero. For each trajectory, The principal comparison between three conditions the distribution of the logarithm of the curve radii is com- shows a significant difference over the whole sessions puted and distributed in three categories (small radii, curvi- (p < 02). As plotted in Figure 6, more collisions occur in linear trajectories, and straight lines), according to a the manual and nonbiological conditions than in the bio- continuous scale of ranges that permits performing a statisti- logical condition. The statistical analysis session by session cal analysis of the results. To finish, we normalized the distri- indicates a significant difference in session 1 (p < 01) and butions, in which the occurrences of radii of curvature in session 4 (p < 03). This last fact suggests that the learning each category are represented by a percentage of all the effect does not enable the users in the manual and nonbi- occurrences for each trajectory. Mean completion time (sec) Mean number of collisions Applied Bionics and Biomechanics 7 mental condition (Figure 9). This advanced analysis of the motor performance shows that the operator tends to max- imize the smoothness of the robot trajectories, when the vehicle replicates the natural human scheme described by the power law. 6. Discussion This study consisted in analyzing the effect of the implemen- tation of the 2/3 power law on the steering control of a vehi- cle. Three experimental conditions were compared. In the Sessions first condition, the participant had to manually control both the speed and the direction of the robot. In the second condi- Manual Biologic tion, the velocity of the vehicle was automatically set accord- Nonbiologic ing to the bioinspired model. Lastly, in the third condition, which was used as a control, the robot speed was automati- Figure 7: Representation of the average rate of large curve radii for cally calculated through an equation that violated the biolog- each of the main conditions (manual vs. biologic vs. nonbiologic) ical motion. The task of the subjects was to remote control against the four experimental sessions. the robot, in order to complete the course as safe and fast as possible. The performance of the participants was recorded on four sessions. The statistical analyses indicate that the number of collisions and the completion time dimin- ish significantly over the sessions. This overall improvement of the performance seems to be related to a learning effect. The main comparison of the study shows that the precision and velocity to accomplish the task are significantly better in the biological condition than in the manual and nonbiolo- gic conditions. Since the speed control is automatic in the biological condition, less sensorimotor resources and mental workload of the teleoperators are required to complete the task. This aspect brings an advantage for the individuals, who can focus their attention on the guidance of the vehicle. Sessions Nevertheless, the fact that the nonbiological condition is sig- Manual nificantly worse than the biological condition means that the Biologic automatic setting of the speed must replicate certain charac- Nonbiologic teristics of the natural movement to be effective. The comparison of the raw performances (speeds and Figure 8: Representation of the average rate of small curve radii for collisions) was complemented by a more advanced assess- each of the main conditions (manual vs. biologic vs. nonbiologic) ment based on the analysis of the robot’s kinematics. The against the four experimental sessions. radii of curvature of the vehicle trajectory were analyzed, in order to evaluate the smoothness of the movements. Like The distribution of large (Figure 7) and small (Figure 8) the raw performance, this parameter shows the benefitof radii of curvatures is not the same whether the subjects inter- implementing a human-like behavior in the robot’s way of act with a robot that implements the human-like behavior or working. The trajectories are significantly smoother when a robot that implements the two other modes of control. the power law is integrated into the robot than when this Thus, the percentage of occurrences of curvilinear trajecto- bioinspired model is absent. Remarkably, the study shows ries is significantly higher in the biologic than in the manual that the advantage of the biological law lasts until the end and nonbiologic conditions (p < 01). Similarly, small radii of the experiment, which supposes a stronger impact of the and turn in place are statistically more frequent in the implementation than the learning effect. This result suggests manual and nonbiological than in the biological condition that the power law and minimum jerk are indeed related to (p < 01). In addition, these significant differences are main- each other. Such an outcome is supported by studies that tained stable over the whole duration of the experiment. It tend to demonstrate that the 2/3 power law is an optimal means that four sessions are not enough to provide the tel- solution to smooth the trajectory, because it sets the normal eoperator with a learning effect that could counterbalance component of the jerk to zero [32, 39, 41]. In addition, it the benefit of the bioinspired semiautonomous mode, in seems that this law satisfies the principle of least action, terms of the rate of both jerky trajectories (p < 01, at ses- which states that the amount of work required to complete sion 4) and smooth movements (p < 01, at session 4). a trajectory is minimal if the movement obeys the 2/3 The difference of steering control can be confirmed by the power law [42]. This observation is consistent with an visualization of the typical paths recorded for each experi- experiment of telemanipulation showing that the motor % of occurrences of small curve radii % of occurrences of large curve radii 8 Applied Bionics and Biomechanics (a) (b) Figure 9: Typical example of a path (green dotted line) performed by a robot controlled in the manual or nonbiological mode (a). Notable is the sharp pattern that occurs before the main changes of direction. Sample of a path performed by a robot controlled in the biological mode (b). This condition is characterized by uniformly smoothed trajectories of the vehicle. skill and performance is negatively correlated with the the mirror neuron system areas is modulated by the mental workload of the surgeon during robot-assisted sur- observer’s motor experience [50, 51]. According to predictive gery [43]. This finding suggests that the smoothness of coding, the optimal state is a minimal prediction error at all levels of the AON, which is achieved when the observed the robot movements controlled by an operator could be used as an indirect measurement of the workload. actions match predicted actions (based on prior visuomotor Furthermore, the fact that the control of a nonanthropo- experience) as closely as possible [52, 53]. To conclude, it is important to mention that it is not morphic robot is significantly improved when the artefact behaves according to the 2/3 power law supports the hypoth- always an advantage to automate some parameters of the esis of the CNS (Central Nervous System) origin of this law artefact in a situation of human-machine interaction. Our [21, 44]. Viviani and Flash [32] described a correlation study suggests that the characteristics of the human being between the power law and movement prediction, in order must be taken into account to create appropriate usability to plan and choose the best trajectory. More precisely, these rules. Here, the proposed method is to implement a bioin- authors underline that the estimation of the trajectory geom- spired behavior to automate the velocity of a robot. In the etry must be accessible to the motor control system as a part case study of the teleoperation of a mobile device or robotic of the internal representation of the predicted movement arms, the anthropocentric approach seems to be efficient. A intention. This is a fundamental feature of the locomotion current trend in the automobile industry is to produce more and more autonomic vehicles [54], which is in a certain sense that requires to program changes in direction one step ahead, in order to overcome the delays due to biomechanical inertia in contradiction with the will of the drivers, who want to keep [45]. This motor coordination seems also to occur during the the control on the technology. Our results suggest that execution of a movement mediated by an artefact, which sug- modeling and implementing human-like behaviors in the gests that this control rule is characteristic of a general machine, such as the two-thirds power law [23] or Fitts’ law [55], is a promising alternative approach for the autom- scheme of the organization of the action. This observation is supported by the replication of the two-thirds power law atization of key processes in the artefact’s way of working. in a mobile robot with quite different (bio)mechanics than The advantage of such a method comes from the fact that a the human being, which would confirm the hypothesis that car behaving as a living being can be easily understood and this law is not dependent on peripheral biomechanic factors appropriated by the end user [36]. Future work will consist in exploring other approaches based on machine learning [30, 46], but as issued from an internal model of the move- ment planning [21, 47]. or reinforcement learning to train the robot to acquire Moreover, the fact that the operator observes a mobile human-like behaviors and, also, improving the transparency device that has human-like kinematics can also explain the of the remote control by providing the operator with natural advantage of the biological mode over the nonbiological. Sev- user interfaces, such as the Kinect, to interact with the machine [17]. eral experiments show that the observation of a biologically plausible movement facilitates the simultaneous execution, by the observer, of a congruent action [2, 48]. Mirror neu- Data Availability rons, and more specifically the Action Observation Network (AON), seem to be involved in this process [49]. In fact, sev- The data used to support the findings of this study are avail- eral neuroimaging studies have shown that the activation of able from the corresponding author upon request. Applied Bionics and Biomechanics 9 [14] M. F. Land and D. N. Lee, “Where we look when we steer?,” Conflicts of Interest Nature, vol. 369, no. 6483, pp. 742–744, 1994. The authors declare that there is no conflict of interest [15] S. B. Shafiei, L. Cavuoto, and K. A. Guru, “Motor skill evalua- regarding the publication of this paper. tion during robot-assisted surgery,” in Proceedings of the 41st Mechanisms and Robotics Conference, Cleveland, Ohio, USA, Acknowledgments [16] E. De Momi, L. Kranendonk, M. Valenti, N. Enayati, and G. Ferrigno, “A neural network-based approach for trajectory This research was funded by the Universidad de Las planning in robot–human handover tasks,” Frontiers in Robot- Américas, Dirección General de Investigación. ics and AI, vol. 3, 2016. [17] H. Su, N. Enayati, L. Vantadori, A. Spinoglio, G. Ferrigno, and References E. De Momi, “Online human-like redundancy optimization for teleoperated anthropomorphic manipulators,” Interna- [1] R. C. Arkin and L. Moshkina, Affect in Human-Robot Interac- tional Journal of Advanced Robotic Systems, vol. 15, no. 6, tion, Georgia Institute of Technology, 2014. [2] A. Kupferberg, S. Glasauer, M. Huber, M. Rickert, A. Knoll, [18] W. Kim, J. Lee, L. Peternel, N. Tsagarakis, and A. Ajoudani, and T. Brandt, “Biological movement increases acceptance of “Anticipatory robot assistance for the prevention of human humanoid robots as human partners in motor interaction,” static joint overloading in human-robot collaboration,” IEEE Experimental Brain Research, vol. 26, no. 4, pp. 339–345, 2011. Robotics and Automation Letters, vol. 3, no. 1, pp. 68–75, 2018. [3] T. Chaminade, D. Franklin, E. Oztop, and G. Cheng, “Motor [19] L. Peternel, N. Tsagarakis, and A. Ajoudani, “A human-robot interference between humans and humanoid robots: effect of co-manipulation approach based on human sensorimotor biological and artificial motion,” in Proceedings of the 4th information,” IEEE Transactions on Neural Systems and Reha- International Conference on Development and Learning, bilitation Engineering, vol. 25, no. 7, pp. 811–822, 2017. pp. 96–101, Osaka, Japan, 2005. [20] M. Huber, M. Rickert, A. Knoll, T. Brandt, and S. Glasauer, [4] E. Oztop, D. Franklin, T. Chaminade, and G. Cheng, “Human- “Human-robot interaction in handing-over tasks,” in Proceed- humanoid interaction: is a humanoid robot perceived as a ings of the 17th IEEE International Symposium on Robot and human?,” in Proceedings of the 4th IEEE International Confer- Human Interactive Communication, pp. 107–112, Munich, ence on Humanoid Robots, pp. 537–559, Santa Monica, CA, Germany, 2008. USA, 2004. [21] F. Lacquaniti, C. Terzuolo, and P. Viviani, “The law relating [5] J. Kilner, A. F. . C. Hamilton, and S. J. Blakemore, “Interference the kinematic and figural aspects of drawing movements,” effect of observed human movement on action is due to veloc- Acta Psychologica, vol. 54, no. 1-3, pp. 115–130, 1983. ity profile of biological motion,” Social Neuroscience, vol. 2, [22] P. Viviani and R. Schneider, “A developmental study of the no. 3-4, pp. 158–166, 2007. relationship between geometry and kinematics in drawing [6] S. Coradeschi, H. Ishiguro, M. Asada et al., “Human-inspired movements,” Journal of Experimental Psychology: Human Per- robots,” IEEE Intelligent Systems, vol. 21, no. 4, pp. 74–85, ception and Performance, vol. 17, no. 1, pp. 198–218, 1991. [23] D. Carvalho, Controlo remoto de um robô com recurso ao sis- [7] B. R. Duffy, “Anthropomorphism and the social robot,” Robot- tema Android [M.S. thesis], Nova University of Lisbon, Lisbon, ics and Autonomous Systems, vol. 42, no. 3-4, pp. 177–190, Portugal, 2015, http://hdl.handle.net/10362/16548. [24] A. M. Jarc and I. Nisky, “Robot-assisted surgery: an emerging [8] Y. Rybarczyk, S. Galerne, P. Hoppenot, E. Colle, and platform for human neuroscience research,” Frontiers in D. Mestre, “The development of robot human-like behaviour Human Neuroscience, vol. 9, 2015. for an efficient human-machine co-operation,” in Proceedings [25] F. Despinoy, D. Bouget, G. Forestier et al., “Unsupervised tra- of the 6th International Conference of the Association for the jectory segmentation for surgical gesture recognition in Advancement of Assistive Technology in Europe, Ljubljana, robotic training,” IEEE Transactions on Biomedical Engineer- Slovenia, 2001. ing, vol. 63, no. 6, pp. 1280–1291, 2016. [9] Y. Rybarczyk, O. Ait Aider, P. Hoppenot, and E. Colle, “Remote control of a biomimetics robot assistance system for [26] Y. Rybarczyk, P. Hoppenot, E. Colle, and D. Mestre, “Sensori- motor appropriation of an artefact: a neuroscientific disabled persons,” AMSE Modelling, Measurement and Con- trol, vol. 63, no. 4, pp. 47–56, 2002. approach,” in Human Machine Interaction—Getting Closer, pp. 187–212, InTechOpen, London, UK, 2012. [10] Y. Rybarczyk, D. Mestre, P. Hoppenot, and E. Colle, “Implé- mentation de mécanismes d’anticipation visuo-motrice en tél- [27] Y. Rybarczyk, D. Mestre, P. Hoppenot, and E. Colle, “A biolog- éopération,” Le Travail Humain, vol. 67, no. 3, pp. 209–233, ical model for the evaluation of human-machine adaptation,” 2004. AMSE Modelling, Measurement and Control, vol. 65, no. 78, pp. 23–33, 2004. [11] Y. P. Rybarczyk and D. Mestre, “Effect of temporal organiza- tion of the visuo-locomotor coupling on the predictive steer- [28] E. T. Levy, K. E. Gordon, and N. P. Fey, “Residual limb revision ing,” Frontiers in Psychology, vol. 3, 2012. surgery alters velocity-curvature coupling during stepping and turning of a transfemoral amputee,” in Proceedings of the 40th [12] R. Grasso, S. Glasauer, Y. Takei, and A. Berthoz, “The predic- tive brain: anticipatory control of head direction for the steer- Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Honolulu, HI, USA, 2018. ing of locomotion,” NeuroReport, vol. 7, no. 6, pp. 1170–1174, [29] S. Schaal and D. Sternad, “Origins and violations of the 2/3 [13] J. P. Wann and R. M. Wilkie, “The role of gaze fixation in loco- power law in rhythmic three-dimensional arm movements,” motor control,” Journal of Vision, vol. 3, no. 9, 2003. Experimental Brain Research, vol. 136, no. 1, pp. 60–72, 2001. 10 Applied Bionics and Biomechanics [46] J. Wann, I. Nimmo-Smith, and A. M. Wing, “Relation between [30] P. L. Gribble and D. J. Ostry, “Origins of the power law relation between movement velocity and curvature: modeling the velocity and curvature in movement: equivalence and diver- effects of muscle mechanics and limb dynamics,” Journal of gence between a power law and a minimum-jerk model,” Jour- Neurophysiology, vol. 76, no. 5, pp. 2853–2860, 1996. nal of Experimental Psychology: Human Perception and Performance, vol. 14, no. 4, pp. 622–637, 1988. [31] M. Zago, A. Matic, T. Flash, A. Gomez-Marin, and F. Lacquaniti, “The speed-curvature power law of movements: [47] J. T. Massey, J. T. Lurito, G. Pellizzer, and A. P. Georgopoulos, a reappraisal,” Experimental Brain Research, vol. 236, no. 1, “Three-dimensional drawings in isometric conditions: relation pp. 69–82, 2018. between geometry and kinematics,” Experimental Brain Research, vol. 88, no. 3, pp. 685–690, 1992. [32] P. Viviani and T. Flash, “Minimum-jerk, two-thirds power law, and isochrony: converging approaches to movement [48] M. Jeannerod, “Neural simulation of action: a unifying mech- planning,” Journal of Experimental Psychology: Human Per- anism for motor cognition,” NeuroImage, vol. 14, no. 1, ception and Performance, vol. 21, no. 1, pp. 32–53, 1995. pp. S103–S109, 2001. [33] D. Huh and T. J. Sejnowski, “Spectrum of power laws for [49] G. Rizzolatti and C. Sinigaglia, “The functional role of the curved hand movements,” Proceedings of the National Acad- parieto-frontal mirror circuit: interpretations and misinterpre- emy of Sciences of the United States of America, vol. 112, tations,” Nature Reviews Neuroscience, vol. 11, no. 4, pp. 264– no. 29, pp. E3950–E3958, 2015. 274, 2010. [34] M. Zago, F. Lacquaniti, and A. Gomez-Marin, “The speed– [50] B. Calvo-Merino, J. Grèzes, D. E. Glaser, R. E. Passingham, and curvature power law in Drosophila larval locomotion,” Biology P. Haggard, “Seeing or doing? Influence of visual and motor Letters, vol. 12, no. 10, 2016. familiarity in action observation,” Current Biology, vol. 16, [35] S. Vieilledent, Y. Kerlirzin, S. Dalbera, and A. Berthoz, no. 19, pp. 1905–1910, 2006. “Relationship between velocity and curvature of a human [51] E. S. Cross, D. J. M. Kraemer, A. F. . C. Hamilton, W. M. locomotor trajectory,” Neuroscience Letters, vol. 305, no. 1, Kelley, and S. T. Grafton, “Sensitivity of the action observa- pp. 65–69, 2001. tion network to physical and observational learning,” Cere- [36] C. de’Sperati and P. Viviani, “The relationship between curva- bral Cortex, vol. 19, no. 2, pp. 315–326, 2009. ture and velocity in two-dimensional smooth pursuit eye [52] A. Neal and J. M. Kilner, “What is simulated in the action movements,” Journal of Neuroscience, vol. 17, no. 10, observation network when we observe actions?,” The Euro- pp. 3932–3945, 1997. pean Journal of Neuroscience, vol. 32, no. 10, pp. 1765–1770, [37] R. S. Marken and D. M. Shaffer, “The power law of movement: an example of a behavioral illusion,” Experimental Brain [53] M. B. Schippers and C. Keysers, “Mapping the flow of informa- Research, vol. 235, no. 6, pp. 1835–1842, 2017. tion within the putative mirror neuron system during gesture [38] R. S. Marken and D. M. Shaffer, “The power law as behavioral observation,” NeuroImage, vol. 57, no. 1, pp. 37–44, 2011. illusion: reappraising the reappraisals,” Experimental Brain [54] K. Curran and J. Lee, “Autonomic vehicles,” Computer and Research, vol. 236, no. 5, pp. 1537–1544, 2018. Information Science, vol. 1, no. 1, 2008. [39] S. B. Shafiei, “Intrinsic constraint applied by central nervous [55] J. V. Drapers, B. C. Jared, and M. W. Noakes, “Manipulator system during motor command creation of smooth hand performance evaluation using Fitts’ taping task,” in Proceed- movements,” Archives of Physical Medicine and Rehabilitation, ings of the 8th Topical Meeting on Robotics and Remote Sys- vol. 97, no. 10, article e49, 2016. tems, Pittsburgh, PA, USA, 1999. [40] P. Péruch and D. Mestre, “Between desktop and head immer- sion: functional visual field during vehicle control and naviga- tion in virtual environments,” Presence: Teleoperators and Virtual Environments, vol. 8, no. 1, pp. 54–64, 1999. [41] E. Todorov and M. I. Jordan, “Smoothness maximization along a predefined path accurately predicts the speed profiles of complex arm movements,” Journal of Neurophysiology, vol. 80, no. 2, pp. 696–714, 1998. [42] S. Lebedev, W. H. Tsui, and P. Van Gelder, “Drawing move- ments as an outcome of the principle of least action,” Journal of Mathematical Psychology, vol. 45, no. 1, pp. 43–52, 2001. [43] K. A. Guru, S. B. Shafiei, A. Khan, A. A. Hussein, M. Sharif, and E. T. Esfahani, “Understanding cognitive performance during robot-assisted surgery,” Urology, vol. 86, no. 4, pp. 751–757, 2015. [44] P. Viviani and N. Stucchi, “Biological movements look uni- form: evidence of motor-perceptual interactions,” Journal of Experimental Psychology: Human Perception and Performance, vol. 18, no. 3, pp. 603–623, 1992. [45] A. E. Patla, S. D. Prentice, C. Robinson, and J. Neufeld, “Visual control of locomotion: strategies for changing direction and for going over obstacles,” Journal of Experimental Psychology: Human Perception and Performance, vol. 17, no. 3, pp. 603– 634, 1991. International Journal of Advances in Rotating Machinery Multimedia Journal of The Scientific Journal of Engineering World Journal Sensors Hindawi Hindawi Publishing Corporation Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Journal of Control Science and Engineering Advances in Civil Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com Journal of Journal of Electrical and Computer Robotics Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 VLSI Design Advances in OptoElectronics International Journal of Modelling & Aerospace International Journal of Simulation Navigation and in Engineering Engineering Observation Hindawi Hindawi Hindawi Hindawi Volume 2018 Volume 2018 Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com www.hindawi.com www.hindawi.com Volume 2018 International Journal of Active and Passive International Journal of Antennas and Advances in Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Bionics and Biomechanics Hindawi Publishing Corporation

Bioinspired Implementation and Assessment of a Remote-Controlled Robot

Loading next page...
 
/lp/hindawi-publishing-corporation/bioinspired-implementation-and-assessment-of-a-remote-controlled-robot-I02Yp2RUu9
Publisher
Hindawi Publishing Corporation
Copyright
Copyright © 2019 Yves Rybarczyk and Diogo Gil Carvalho. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ISSN
1176-2322
eISSN
1754-2103
DOI
10.1155/2019/8575607
Publisher site
See Article on Publisher Site

Abstract

Hindawi Applied Bionics and Biomechanics Volume 2019, Article ID 8575607, 10 pages https://doi.org/10.1155/2019/8575607 Research Article Bioinspired Implementation and Assessment of a Remote- Controlled Robot 1,2 3 Yves Rybarczyk and Diogo Gil Carvalho Intelligent & Interactive Systems Lab (SI2 Lab), Universidad de Las Américas, 170125 Quito, Ecuador Dalarna University, 791-88 Falun, Sweden Department of Electrical Engineering, CTS/UNINOVA, Nova University of Lisbon, 2829-516 Monte de Caparica, Portugal Correspondence should be addressed to Yves Rybarczyk; y.rybarczyk@fct.unl.pt Received 26 February 2019; Revised 9 May 2019; Accepted 21 August 2019; Published 11 September 2019 Guest Editor: Francesca Cordella Copyright © 2019 Yves Rybarczyk and Diogo Gil Carvalho. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Daily activities are characterized by an increasing interaction with smart machines that present a certain level of autonomy. However, the intelligence of such electronic devices is not always transparent for the end user. This study is aimed at assessing the quality of the remote control of a mobile robot whether the artefact exhibits a human-like behavior or not. The bioinspired behavior implemented in the robot is the well-described two-thirds power law. The performance of participants who teleoperate the semiautonomous vehicle implementing the biological law is compared to a manual and nonbiological mode of control. The results show that the time required to complete the path and the number of collisions with obstacles are significantly lower in the biological condition than in the two other conditions. Also, the highest percentage of occurrences of curvilinear or smooth trajectories are obtained when the steering is assisted by an integration of the power law in the robot’s way of working. This advanced analysis of the performance based on the naturalness of the movement kinematics provides a refined evaluation of the quality of the Human-Machine Interaction (HMI). This finding is consistent with the hypothesis of a relationship between the power law and jerk minimization. In addition, the outcome of this study supports the theory of a CNS origin of the power law. The discussion addresses the implications of the anthropocentric approach to enhance the HMI. 1. Introduction the robot seems relevant. For instance, the human-robot interaction tends to be improved when the machine has a Industries face an increasing demand for collaborative robots humanoid appearance [6]. This fact can be explained by an inconscient tendency of the human being to anthropomor- that exhibit human-like behaviors. This trend is justified by the fact that it is easier for an operator to predict the actions phize the artefacts they interact with, in order to predict their of a robot that behaves more like a human being than like a behavior and increase their acceptance of the machine [7]. machine [1]. A study that uses the experimental paradigm Besides the situation of interaction, the implementation of Motor Interference (MI) shows that the motor perfor- of human-like behaviors in a robot’s way of working also mance of an individual can be influenced by the perception seems to benefit an operator that has to control a machine. of the movements of a robot, if the machine replicates some This statement is particularly true in the context of teleopera- characteristics of biological motion [2–4]. In particular, it tion, which implies several limitations for a human operator. seems that the movement velocity profile is sufficient to cre- For instance, the sensorial information received by the teleo- ate this interference. This result suggests that a movement perator can be altered, for example, the field of view is can be processed as biologic by the human brain, even if it reduced, not all the sensorial modalities are restituted (e.g., is not produced by a living being, on the condition that the audition and proprioception), and the response of the system artefact motion simulates (even approximatively) certain is delayed. Another aspect is the necessity to build or accom- biological kinematics [5]. Moreover, the physical aspect of modate new motor schemes to be able to control the user 2 Applied Bionics and Biomechanics aimed at assessing the motor control of a robot arm to interface of the device, which augments the mental workload. A promising approach to reduce the gap between the user assist surgeons [15]. The results show that both smooth- and the telerobot is to implement human-like behaviors in ness and minimum jerk are significant measures of exper- a robot [8, 9]. For instance, Rybarczyk et al. [10, 11] have tise levels. The end-effector trajectory evolves from sharp studied the effect of the implementation in a mobile robot and jerky in novices to smooth in experts. Thus, the of the human behavior of visuomotor anticipation over the authors conclude that these two features are excellent cri- locomotion, in which the direction of the robot pan-tilt teria to evaluate motor skill in the conditions of human- camera is automatically oriented toward the tangent point robot interaction. Although the power law is also identi- of the inside curve of the path, as walkers/cyclists/drivers fied as a discriminant measure of expertise, registering do [12–14]. The results show that the motor performances such a biological law seems to depend on the characteris- tics of the artefact. For example, some studies have dem- of the teleoperators are enhanced when they steer the bioinspired robot. A correlation between the replication of onstrated that this law is replicated in situations of biological laws and the level of expertise is also observed in teleoperation [11] and use of prostheses [28]. the case of the telemanipulation of robotic arms, such as in Actually, there is a controversy regarding the origins telesurgery [15]. and the violations of the 2/3 power law during the execu- Different strategies are used to implement human-like tion of the biological movements [22, 29–31]. On the one behaviors in a robot. A traditional approach applied in the hand, some studies tend to demonstrate that the power industry is to create anthropomorphic collaborative robots law is a signature of the Central Nervous System (CNS) (or cobots) that are trained to imitate biological motions, [32–34], because it seems to be independent of the dynam- through machine learning algorithms [16]. In the case of ics of the limbs. This law is indeed observed in a wide vari- the teleoperation, it seems that individuals feel also more ety of activities such as drawing [21], walking [35], and comfortable to control an anthropomorphic robot arm in smooth pursuit eye [36]. On the other hand, different stud- which the motion trajectory of the end effector is like a bio- ies defend a biomechanical [30] or, even, an artefactual logical movement [17]. Jerk minimization is one of the prin- explanation [37, 38]. There are also contradictory results cipal human-like behaviors that has been implemented to regarding the relationship between smoothness, minimum model a natural trajectory planning [18, 19]. The minimum jerk, and power law. Some studies show evidences that jerk is characterized by a bell-shaped velocity profile, in these features are related to each other [32, 39], whereas which the movement speed increases progressively, reaches others suggest the contrary [15, 29]. a peak near the midpoint, and then deceases slowly. This The present work attempts to tackle these different absence of abrupt changes seems to support the execution contradictory findings about the 2/3 power law by inte- of a smooth motion [20]. Another fundamental motor grating this bioinspired kinematics in a remote-controlled behavior is the relationship between the velocity and the mobile robot. An experiment is designed to compare the curvature of the biological movements, which is known teleoperation of a robot with the 2/3 power law (biological as the two-thirds power law [21, 22]. This law states that condition) versus two modes of control that do not imple- the angular velocity of the end effector is proportional to ment this human-like behavior (manual condition and the two-third root of its curvature or, equivalently, that nonbiological condition). In the biological mode, the the instantaneous tangential velocity (v ) is proportional engine speed is automatically servo controlled by the vehi- to the third root of the radius of curvature (r ), as cle trajectory according to the power law equation. In the manual mode, the user has to control both the velocity described in equation (1). In other words, it means that the velocity of the movement decreases in the highly and the direction of the mobile device. In the third condi- curved parts of the trajectory and increases when the tra- tion, the vehicle speed is also automatic, but the calcula- jectory becomes straighter. Implementing this model in a tion of the relationship between geometry and kinematics mobile robot tends to improve the raw performance when violates the biological motion. This last condition is used as a control to make sure that the potential difference of steering the vehicle [23]. performance between the two main conditions (biologic −1/3 vs. manual) is not caused by a dissimilar complexity of v = kr 1 t t the task (i.e., number of parameters that must be con- Nevertheless, few studies are interested in considering trolled by the participants). We posit the hypothesis that semiautonomous driving, in which the velocity is automat- refined features to gauge the quality of the Human- Machine Interaction (HMI). Instead of focusing only on ically set according to the power law principles (biological the raw performance (e.g., completion time of the task mode), should promote a significantly faster, safer, and and percentage occurrence of errors), these studies analyze more natural steering than the nonassisted (manual mode) the kinematics of the robot control [24–26]. To proceed and nonbiologic (artificial mode) control. The quality of the interaction is assessed from both the raw performance with such an advanced assessment, the human behaviors are now used as criteria to estimate an appropriate inter- (completion time and number of collisions) and refined parameters based on the smoothness of the trajectories. action. For instance, minimum jerk, smoothness, and 2/3 power law can be applied as a reference to evaluate a suit- The remainder of the manuscript is organized into able interaction between a human operator and an artefact three main sections. First, the implementation of the tele- operation system is described. The experimental protocol [17, 27]. These three features are compared in a study that Applied Bionics and Biomechanics 3 Motor Motor Motor (a) (b) Figure 1: (a) Schematic representation of the robot (top view) designed for the study. Two independent motors drive the front wheels, and a third one controls the rotation of the pan camera. This mobile vision is implemented by default to promote a visual anticipation over the change of direction. (b) Illustration of the camera behavior in some specific locations of the path. The blue arrow indicates the instantaneous direction of the vehicle and the red arrow represents the orientation of the pan camera at the same moment. It is notable that the angle between the two arrows is inversely proportional to the radius of curvature of the robot trajectory. The more curved is the shape of the path (e.g., position 2), the larger is the angle between the orientation of the camera and the direction of the vehicle, and vice versa (e.g., position 4). 2.2. Robot Behavior. The vehicle is built on four wheels, and conditions (manual vs. nonbiologic vs. biologic) are also explained in detail. Second, the results of the perfor- employing a front-wheel-drive system (Figure 1(a)). The mance for each condition are presented, analyzed, and dis- two front wheels are moved by two independent motors. cussed. Finally, the outcomes are interpreted, in order to The differential of speed between the right and the left draw some conclusions and perspectives regarding the appli- wheel rotation allows the vehicle to turn. The pan camera is set on a mobile structure, which is moved by another cation of the anthropocentric approach in the human-robot interaction, as well as the origins of the power law and its motor. The orientation of the camera is determined auto- relationship with jerk minimization. matically based on the direction of the robot, that is, the camera points toward the inside of the vehicle trajectory. Since any change of direction is systematically anticipated 2. Material and Methods by a rotation of the camera proportional to the curvature of the vehicle trajectory, a visual prediction over the robot 2.1. System Architecture. The three main elements that com- motion is provided to the operator. This mechanism pose the system are (i) a NXT mobile robot, (ii) an Android inspired from the human behavior [12, 14] is implemented device for the remote control, and a pan IP camera. Since by default, because it facilitates the teleoperation [8, 10]. the experiment is carried out in a situation of teleopera- Figure 1(b) shows examples of this visuolocomotor cou- tion (i.e., indirect perception and action on the robot envi- pling between camera and robot for different curves of ronment), a wireless connection is used to support the the path. communication between the principal components of the architecture. Two different protocols of communication 3. Experimental Conditions are applied. The Android-based remote control communi- cates with the NXT through Bluetooth technology. In 3.1. Manual Condition. Both speed and direction of the addition, the connection between the IP camera and the vehicle are manually controlled by the operator in this smartphone is supported by Wi-Fi communication. The experimental condition. Concentric semicircles that corre- robot is connected to the IP camera thanks to a support spond to different speed levels are displayed on the control library that permits the system integration between the panel of the user’s interface (Figure 2(a)). The bigger is the two entities. Thus, the operators use the Android remote radius of the semicircle, the higher is the speed. Thus, the control device to interact with the whole system, which vehicle velocity is calculated based on the distance between allows them to steer the mobile robot and receive a visual the center of all concentric semicircles and the selected feedback from the pan IP camera. An Android application semicircle. The direction of the robot is determined by is developed and implemented on the smartphone to per- the angle between the vertical of the screen and the loca- mit such an interaction. The tactile user interface enables tion of the user’s fingertip. The range of angles goes from ° ° the operator to control the trajectories of the vehicle, to 0 to 180 , rotating counter clockwise. If the fingertip of ° ° choose the steering mode of the robot (manual vs. nonbio- the user is positioned between 0 and 90 , the robot turns logic vs. biologic), to calibrate the pan camera, and to turn right, with a curvature proportional to the angle between the system on or off. the vertical (90 ) and the position of the finger (the more 4 Applied Bionics and Biomechanics (a) (b) Figure 2: (a) Representation of the GUI for the manual mode of driving. Each concentric circle represents a different speed (the larger the radius of the semicircle, the higher the velocity). (b) User interface for the biological and nonbiological conditions. A single semicircle enables the user to directly control the direction of the robot and indirectly set the speed of the vehicle. the location of the finger tends to 0 , the more the vehicle angle should perfectly fit to the human’s skills. On the turns right). On the contrary, if the position of the finger contrary, because of its unnatural behavior, the nonbiolog- ° ° is between 90 and 180 , the vehicle turns left (again, the ical semiautonomous control should be more challenging radius of the curvature of the trajectory depends on the for the teleoperator. Figure 3 summarizes, through a block diagram, the differences between the manual and semiau- angle from the vertical of the screen). The controller of the robot is constantly waiting for an input sent from tomatic modes of control. the graphic user interface, in order to update the direction and speed of the mobile platform. 4. Experimental Protocol 3.2. Biological and Nonbiological Conditions. In these driv- Thirty people (15 males and 15 females; 23 5± 3 5 years) ing modes, the user has only to use the touchscreen inter- took part in the experiment. All the participants had a nor- face to control the trajectory of the robot. The speed is mal or corrected-to-normal vision. The procedure con- automatically set according to the direction of the vehicle. formed to the Declaration of Helsinki and was approved In the biological condition, the 2/3 power law is used to by the Ethical Review Board of the Nova University of calculate the speed, which is based on the instantaneous Lisbon. The experiment was carried out in a classroom, radius of the curvature of the robot trajectory. The maxi- where the subjects had to teleoperate the NXT vehicle mum velocity of the robot is 30 cm/s, if the vehicle goes through the Android-based mobile device. The instructions straight forward. In the case that the radius of curvature provided to the participants were to steer the robot as safe decreases (to the right or to the left), the robot’s speed (the least collisions) and fast (the minimum completion diminishes by a rate of one-third (see equation (1)). In time) as possible through a path delimited by plastic the nonbiological condition, the velocity of the vehicle is blocks. The entire distance of the route was approximately also automatic, but it is not set according to the biological seven meters and consisted of numerous bends and motion. The relationship between speed and geometry changes in direction (curves and countercurves). The does not follow a power law, but a linear law described in sequence of the course was as follows: (i) a straight line, ° ° (ii) an approximately 150 bend, (iii) a 90 reverse curve, v = kr 2 t t (iv) another 150 bend, and (v) a final straight line (Figure 4). A blue adhesive strip marked the starting and Since it is not necessary to modulate manually the finishing line. The symmetric shape of the setup was espe- velocity, the graphic user interface is represented only by cially designed to carry out the route in both directions, a single semicircle (Figure 2(b)). The semicircle allows clockwise and anticlockwise. the operator to control the trajectory of the robot. From After a training session, all the subjects had to execute the user’s perspective, the way to steer the vehicle is iden- the trial twelve times: four repetitions in the manual tical to the manual mode of driving. The user has to inter- mode, four repetitions in the biological mode, and four act with the left and right portion of the semicircle to turn repetitions in the nonbiological mode. The order of the left and right, respectively. The more the fingertip is experimental conditions was counterbalanced from one located to the extremities of the semicircle, the more the subject to another so that ten individuals started with robot turns sharply. The only difference between these the manual control, ten others started with the biological two semiautomatic modes and the manual one is the fact control, and the last ten started with the nonbiological that the velocity is indirectly and automatically set when control. This counterbalancing was implemented to pre- the user chooses a determined direction. Precisely, the vent a possible learning effect, which would bias the out- robot speed is proportional to the selected steering angle. come of the study. For each of the principal conditions Thus, if the power law is adapted to the remote control (manual vs. biologic vs. nonbiologic), the trial was per- of an artefact, the matching between speed and steering formed twice clockwise and twice anticlockwise. The Applied Bionics and Biomechanics 5 Manual Semiautomatic Operator mode modes Visual n feedback Direction (휃) Linear Direction d speed (휈) (휃) Linear speed (휈) 휈 =f(휃 ) O t t Bluetooth Wi-Fi transmission protocol (IP camera) Microcontroller Motor Camera driver motor Left wheel Right wheel motor motor Vehicle motion Figure 3: Block diagram of the two modes of remote control. If the operator picks the manual mode (left side), the speed and direction are controlled independently. On the contrary, if a semiautomatic mode is selected (right side), the robot speed is automatically calculated from the power law (biological condition) or linear law (nonbiological condition) function of the direction defined by the user. completion time, the number of collisions, and the robot ishing line diminishes significantly from session 1 to ses- trajectory were recorded at the end of each trial. sion 4. No interaction effects are detected between the sessions (1, 2, 3, and 4) and the main conditions (manual, biologic, and nonbiologic). 5. Results In addition, the comparison of the completion time between the three conditions indicates a significant differ- The experimental data are statistically analyzed through ANOVA tests for multivariable comparisons and t-tests for ence (p < 005). As shown in Figure 5, the participants complete the task faster in the biological mode than in the pairwise comparisons. the manual (p < 01) and nonbiological (p < 005) steering 5.1. Completion Time. We first analyzed the time perfor- modes. The pairwise analyses confirm the significant dif- ference in session 1 (p < 01), session 2 (p < 05), and session mance of the participants to complete the task. Results indi- cate that the completion time is significantly affected by 3(p < 04). Nevertheless, this statistical difference vanishes in session 4, although the manual and nonbiological modes the experimental sessions (p < 05). A pairwise analysis shows a significant difference between session 1 and ses- tend to remain slower than the biological. The reduction of sion 4 (p < 03). This outcome indicates that the required the completion time over the sessions can be explained by a learning effect that occurs in all the conditions. time to guide the vehicle from the starting line to the fin- 6 Applied Bionics and Biomechanics 5.5 4.5 3.5 2.5 1.5 Sessions Figure 4: Picture of the experimental setting. The symmetric form Manual of the path was chosen to easily alternate the course direction of Biologic the robot from one trial to the next: once clockwise and once Nonbiologic counter clockwise. This alternation was designed to minimize the environment learning and a consequent machine-like driving ° Figure 6: Representation of the average number of collisions for of the vehicle. The two straight lines, two 150 bends, and one each of the main conditions (manual vs. biologic vs. nonbiologic) 90 bend are identified by broken yellow, green, and magenta against the four experimental sessions. lines, respectively. Note that these colors are added for a better understanding of the setup but were not visible during the experiment. ological modes to get steering skills as good as in the bio- logical condition. 5.3. Trajectory Smoothness. The last results address the ques- tion of the movement kinematics through the analysis of the jerk in the control of the robot trajectory. One way to quan- 42 tify the path smoothness is to calculate the instantaneous radius of curvature of each trajectory, then to evaluate the distribution frequency of the radius for all trials [40]. More specifically, the curve radius (r) is computed from the instan- taneous linear velocity (v) divided by the instantaneous rota- tion speed (w), as described in Sessions m/s r = 3 Manual radians/s Biologic Nonbiologic Subsequently, the radius of curvature is converted into a Figure 5: Representation of the mean completion time (in seconds) decimal logarithm. Therefore, if the vehicle has a low linear for each of the main conditions (manual vs. biologic vs. speed and a high velocity of rotation, the curve radius is very nonbiologic) against the four experimental sessions. small (<2), and gets smaller as the velocity of rotation increases. The result is a logarithmic value of r that is around 5.2. Number of Collisions. The assessment of the rate of colli- zero. Conversely, if the vehicle combines a translation and a sions was also performed to complement the analysis. The rotation (curvilinear trajectory), the curve radius is high statistical results indicate that the mean number of collisions (≥2) and its logarithm becomes superior to zero. A steering is significantly different over the sessions (p < 03). The pair- control in which the subject stops and turns in place provides wise analysis shows a significant diminution of the collisions a bimodal distribution of the curve radii, with one peak cen- from session 1 to session 4 (p < 02). These outcomes point tered on null values of the logarithm and another peak cen- out that the subjects have improved the quality of their driv- tered on positive values. On the contrary, a curvilinear (or ing skills over the experiment. There is no effect of interac- smooth) trajectory is characterized by a unimodal pattern tion between the four sessions and the main experimental of distribution centered on a value of the logarithm of the conditions (manual, biologic, and nonbiologic). radius of curvature higher than zero. For each trajectory, The principal comparison between three conditions the distribution of the logarithm of the curve radii is com- shows a significant difference over the whole sessions puted and distributed in three categories (small radii, curvi- (p < 02). As plotted in Figure 6, more collisions occur in linear trajectories, and straight lines), according to a the manual and nonbiological conditions than in the bio- continuous scale of ranges that permits performing a statisti- logical condition. The statistical analysis session by session cal analysis of the results. To finish, we normalized the distri- indicates a significant difference in session 1 (p < 01) and butions, in which the occurrences of radii of curvature in session 4 (p < 03). This last fact suggests that the learning each category are represented by a percentage of all the effect does not enable the users in the manual and nonbi- occurrences for each trajectory. Mean completion time (sec) Mean number of collisions Applied Bionics and Biomechanics 7 mental condition (Figure 9). This advanced analysis of the motor performance shows that the operator tends to max- imize the smoothness of the robot trajectories, when the vehicle replicates the natural human scheme described by the power law. 6. Discussion This study consisted in analyzing the effect of the implemen- tation of the 2/3 power law on the steering control of a vehi- cle. Three experimental conditions were compared. In the Sessions first condition, the participant had to manually control both the speed and the direction of the robot. In the second condi- Manual Biologic tion, the velocity of the vehicle was automatically set accord- Nonbiologic ing to the bioinspired model. Lastly, in the third condition, which was used as a control, the robot speed was automati- Figure 7: Representation of the average rate of large curve radii for cally calculated through an equation that violated the biolog- each of the main conditions (manual vs. biologic vs. nonbiologic) ical motion. The task of the subjects was to remote control against the four experimental sessions. the robot, in order to complete the course as safe and fast as possible. The performance of the participants was recorded on four sessions. The statistical analyses indicate that the number of collisions and the completion time dimin- ish significantly over the sessions. This overall improvement of the performance seems to be related to a learning effect. The main comparison of the study shows that the precision and velocity to accomplish the task are significantly better in the biological condition than in the manual and nonbiolo- gic conditions. Since the speed control is automatic in the biological condition, less sensorimotor resources and mental workload of the teleoperators are required to complete the task. This aspect brings an advantage for the individuals, who can focus their attention on the guidance of the vehicle. Sessions Nevertheless, the fact that the nonbiological condition is sig- Manual nificantly worse than the biological condition means that the Biologic automatic setting of the speed must replicate certain charac- Nonbiologic teristics of the natural movement to be effective. The comparison of the raw performances (speeds and Figure 8: Representation of the average rate of small curve radii for collisions) was complemented by a more advanced assess- each of the main conditions (manual vs. biologic vs. nonbiologic) ment based on the analysis of the robot’s kinematics. The against the four experimental sessions. radii of curvature of the vehicle trajectory were analyzed, in order to evaluate the smoothness of the movements. Like The distribution of large (Figure 7) and small (Figure 8) the raw performance, this parameter shows the benefitof radii of curvatures is not the same whether the subjects inter- implementing a human-like behavior in the robot’s way of act with a robot that implements the human-like behavior or working. The trajectories are significantly smoother when a robot that implements the two other modes of control. the power law is integrated into the robot than when this Thus, the percentage of occurrences of curvilinear trajecto- bioinspired model is absent. Remarkably, the study shows ries is significantly higher in the biologic than in the manual that the advantage of the biological law lasts until the end and nonbiologic conditions (p < 01). Similarly, small radii of the experiment, which supposes a stronger impact of the and turn in place are statistically more frequent in the implementation than the learning effect. This result suggests manual and nonbiological than in the biological condition that the power law and minimum jerk are indeed related to (p < 01). In addition, these significant differences are main- each other. Such an outcome is supported by studies that tained stable over the whole duration of the experiment. It tend to demonstrate that the 2/3 power law is an optimal means that four sessions are not enough to provide the tel- solution to smooth the trajectory, because it sets the normal eoperator with a learning effect that could counterbalance component of the jerk to zero [32, 39, 41]. In addition, it the benefit of the bioinspired semiautonomous mode, in seems that this law satisfies the principle of least action, terms of the rate of both jerky trajectories (p < 01, at ses- which states that the amount of work required to complete sion 4) and smooth movements (p < 01, at session 4). a trajectory is minimal if the movement obeys the 2/3 The difference of steering control can be confirmed by the power law [42]. This observation is consistent with an visualization of the typical paths recorded for each experi- experiment of telemanipulation showing that the motor % of occurrences of small curve radii % of occurrences of large curve radii 8 Applied Bionics and Biomechanics (a) (b) Figure 9: Typical example of a path (green dotted line) performed by a robot controlled in the manual or nonbiological mode (a). Notable is the sharp pattern that occurs before the main changes of direction. Sample of a path performed by a robot controlled in the biological mode (b). This condition is characterized by uniformly smoothed trajectories of the vehicle. skill and performance is negatively correlated with the the mirror neuron system areas is modulated by the mental workload of the surgeon during robot-assisted sur- observer’s motor experience [50, 51]. According to predictive gery [43]. This finding suggests that the smoothness of coding, the optimal state is a minimal prediction error at all levels of the AON, which is achieved when the observed the robot movements controlled by an operator could be used as an indirect measurement of the workload. actions match predicted actions (based on prior visuomotor Furthermore, the fact that the control of a nonanthropo- experience) as closely as possible [52, 53]. To conclude, it is important to mention that it is not morphic robot is significantly improved when the artefact behaves according to the 2/3 power law supports the hypoth- always an advantage to automate some parameters of the esis of the CNS (Central Nervous System) origin of this law artefact in a situation of human-machine interaction. Our [21, 44]. Viviani and Flash [32] described a correlation study suggests that the characteristics of the human being between the power law and movement prediction, in order must be taken into account to create appropriate usability to plan and choose the best trajectory. More precisely, these rules. Here, the proposed method is to implement a bioin- authors underline that the estimation of the trajectory geom- spired behavior to automate the velocity of a robot. In the etry must be accessible to the motor control system as a part case study of the teleoperation of a mobile device or robotic of the internal representation of the predicted movement arms, the anthropocentric approach seems to be efficient. A intention. This is a fundamental feature of the locomotion current trend in the automobile industry is to produce more and more autonomic vehicles [54], which is in a certain sense that requires to program changes in direction one step ahead, in order to overcome the delays due to biomechanical inertia in contradiction with the will of the drivers, who want to keep [45]. This motor coordination seems also to occur during the the control on the technology. Our results suggest that execution of a movement mediated by an artefact, which sug- modeling and implementing human-like behaviors in the gests that this control rule is characteristic of a general machine, such as the two-thirds power law [23] or Fitts’ law [55], is a promising alternative approach for the autom- scheme of the organization of the action. This observation is supported by the replication of the two-thirds power law atization of key processes in the artefact’s way of working. in a mobile robot with quite different (bio)mechanics than The advantage of such a method comes from the fact that a the human being, which would confirm the hypothesis that car behaving as a living being can be easily understood and this law is not dependent on peripheral biomechanic factors appropriated by the end user [36]. Future work will consist in exploring other approaches based on machine learning [30, 46], but as issued from an internal model of the move- ment planning [21, 47]. or reinforcement learning to train the robot to acquire Moreover, the fact that the operator observes a mobile human-like behaviors and, also, improving the transparency device that has human-like kinematics can also explain the of the remote control by providing the operator with natural advantage of the biological mode over the nonbiological. Sev- user interfaces, such as the Kinect, to interact with the machine [17]. eral experiments show that the observation of a biologically plausible movement facilitates the simultaneous execution, by the observer, of a congruent action [2, 48]. Mirror neu- Data Availability rons, and more specifically the Action Observation Network (AON), seem to be involved in this process [49]. In fact, sev- The data used to support the findings of this study are avail- eral neuroimaging studies have shown that the activation of able from the corresponding author upon request. Applied Bionics and Biomechanics 9 [14] M. F. Land and D. N. Lee, “Where we look when we steer?,” Conflicts of Interest Nature, vol. 369, no. 6483, pp. 742–744, 1994. The authors declare that there is no conflict of interest [15] S. B. Shafiei, L. Cavuoto, and K. A. Guru, “Motor skill evalua- regarding the publication of this paper. tion during robot-assisted surgery,” in Proceedings of the 41st Mechanisms and Robotics Conference, Cleveland, Ohio, USA, Acknowledgments [16] E. De Momi, L. Kranendonk, M. Valenti, N. Enayati, and G. Ferrigno, “A neural network-based approach for trajectory This research was funded by the Universidad de Las planning in robot–human handover tasks,” Frontiers in Robot- Américas, Dirección General de Investigación. ics and AI, vol. 3, 2016. [17] H. Su, N. Enayati, L. Vantadori, A. Spinoglio, G. Ferrigno, and References E. De Momi, “Online human-like redundancy optimization for teleoperated anthropomorphic manipulators,” Interna- [1] R. C. Arkin and L. Moshkina, Affect in Human-Robot Interac- tional Journal of Advanced Robotic Systems, vol. 15, no. 6, tion, Georgia Institute of Technology, 2014. [2] A. Kupferberg, S. Glasauer, M. Huber, M. Rickert, A. Knoll, [18] W. Kim, J. Lee, L. Peternel, N. Tsagarakis, and A. Ajoudani, and T. Brandt, “Biological movement increases acceptance of “Anticipatory robot assistance for the prevention of human humanoid robots as human partners in motor interaction,” static joint overloading in human-robot collaboration,” IEEE Experimental Brain Research, vol. 26, no. 4, pp. 339–345, 2011. Robotics and Automation Letters, vol. 3, no. 1, pp. 68–75, 2018. [3] T. Chaminade, D. Franklin, E. Oztop, and G. Cheng, “Motor [19] L. Peternel, N. Tsagarakis, and A. Ajoudani, “A human-robot interference between humans and humanoid robots: effect of co-manipulation approach based on human sensorimotor biological and artificial motion,” in Proceedings of the 4th information,” IEEE Transactions on Neural Systems and Reha- International Conference on Development and Learning, bilitation Engineering, vol. 25, no. 7, pp. 811–822, 2017. pp. 96–101, Osaka, Japan, 2005. [20] M. Huber, M. Rickert, A. Knoll, T. Brandt, and S. Glasauer, [4] E. Oztop, D. Franklin, T. Chaminade, and G. Cheng, “Human- “Human-robot interaction in handing-over tasks,” in Proceed- humanoid interaction: is a humanoid robot perceived as a ings of the 17th IEEE International Symposium on Robot and human?,” in Proceedings of the 4th IEEE International Confer- Human Interactive Communication, pp. 107–112, Munich, ence on Humanoid Robots, pp. 537–559, Santa Monica, CA, Germany, 2008. USA, 2004. [21] F. Lacquaniti, C. Terzuolo, and P. Viviani, “The law relating [5] J. Kilner, A. F. . C. Hamilton, and S. J. Blakemore, “Interference the kinematic and figural aspects of drawing movements,” effect of observed human movement on action is due to veloc- Acta Psychologica, vol. 54, no. 1-3, pp. 115–130, 1983. ity profile of biological motion,” Social Neuroscience, vol. 2, [22] P. Viviani and R. Schneider, “A developmental study of the no. 3-4, pp. 158–166, 2007. relationship between geometry and kinematics in drawing [6] S. Coradeschi, H. Ishiguro, M. Asada et al., “Human-inspired movements,” Journal of Experimental Psychology: Human Per- robots,” IEEE Intelligent Systems, vol. 21, no. 4, pp. 74–85, ception and Performance, vol. 17, no. 1, pp. 198–218, 1991. [23] D. Carvalho, Controlo remoto de um robô com recurso ao sis- [7] B. R. Duffy, “Anthropomorphism and the social robot,” Robot- tema Android [M.S. thesis], Nova University of Lisbon, Lisbon, ics and Autonomous Systems, vol. 42, no. 3-4, pp. 177–190, Portugal, 2015, http://hdl.handle.net/10362/16548. [24] A. M. Jarc and I. Nisky, “Robot-assisted surgery: an emerging [8] Y. Rybarczyk, S. Galerne, P. Hoppenot, E. Colle, and platform for human neuroscience research,” Frontiers in D. Mestre, “The development of robot human-like behaviour Human Neuroscience, vol. 9, 2015. for an efficient human-machine co-operation,” in Proceedings [25] F. Despinoy, D. Bouget, G. Forestier et al., “Unsupervised tra- of the 6th International Conference of the Association for the jectory segmentation for surgical gesture recognition in Advancement of Assistive Technology in Europe, Ljubljana, robotic training,” IEEE Transactions on Biomedical Engineer- Slovenia, 2001. ing, vol. 63, no. 6, pp. 1280–1291, 2016. [9] Y. Rybarczyk, O. Ait Aider, P. Hoppenot, and E. Colle, “Remote control of a biomimetics robot assistance system for [26] Y. Rybarczyk, P. Hoppenot, E. Colle, and D. Mestre, “Sensori- motor appropriation of an artefact: a neuroscientific disabled persons,” AMSE Modelling, Measurement and Con- trol, vol. 63, no. 4, pp. 47–56, 2002. approach,” in Human Machine Interaction—Getting Closer, pp. 187–212, InTechOpen, London, UK, 2012. [10] Y. Rybarczyk, D. Mestre, P. Hoppenot, and E. Colle, “Implé- mentation de mécanismes d’anticipation visuo-motrice en tél- [27] Y. Rybarczyk, D. Mestre, P. Hoppenot, and E. Colle, “A biolog- éopération,” Le Travail Humain, vol. 67, no. 3, pp. 209–233, ical model for the evaluation of human-machine adaptation,” 2004. AMSE Modelling, Measurement and Control, vol. 65, no. 78, pp. 23–33, 2004. [11] Y. P. Rybarczyk and D. Mestre, “Effect of temporal organiza- tion of the visuo-locomotor coupling on the predictive steer- [28] E. T. Levy, K. E. Gordon, and N. P. Fey, “Residual limb revision ing,” Frontiers in Psychology, vol. 3, 2012. surgery alters velocity-curvature coupling during stepping and turning of a transfemoral amputee,” in Proceedings of the 40th [12] R. Grasso, S. Glasauer, Y. Takei, and A. Berthoz, “The predic- tive brain: anticipatory control of head direction for the steer- Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Honolulu, HI, USA, 2018. ing of locomotion,” NeuroReport, vol. 7, no. 6, pp. 1170–1174, [29] S. Schaal and D. Sternad, “Origins and violations of the 2/3 [13] J. P. Wann and R. M. Wilkie, “The role of gaze fixation in loco- power law in rhythmic three-dimensional arm movements,” motor control,” Journal of Vision, vol. 3, no. 9, 2003. Experimental Brain Research, vol. 136, no. 1, pp. 60–72, 2001. 10 Applied Bionics and Biomechanics [46] J. Wann, I. Nimmo-Smith, and A. M. Wing, “Relation between [30] P. L. Gribble and D. J. Ostry, “Origins of the power law relation between movement velocity and curvature: modeling the velocity and curvature in movement: equivalence and diver- effects of muscle mechanics and limb dynamics,” Journal of gence between a power law and a minimum-jerk model,” Jour- Neurophysiology, vol. 76, no. 5, pp. 2853–2860, 1996. nal of Experimental Psychology: Human Perception and Performance, vol. 14, no. 4, pp. 622–637, 1988. [31] M. Zago, A. Matic, T. Flash, A. Gomez-Marin, and F. Lacquaniti, “The speed-curvature power law of movements: [47] J. T. Massey, J. T. Lurito, G. Pellizzer, and A. P. Georgopoulos, a reappraisal,” Experimental Brain Research, vol. 236, no. 1, “Three-dimensional drawings in isometric conditions: relation pp. 69–82, 2018. between geometry and kinematics,” Experimental Brain Research, vol. 88, no. 3, pp. 685–690, 1992. [32] P. Viviani and T. Flash, “Minimum-jerk, two-thirds power law, and isochrony: converging approaches to movement [48] M. Jeannerod, “Neural simulation of action: a unifying mech- planning,” Journal of Experimental Psychology: Human Per- anism for motor cognition,” NeuroImage, vol. 14, no. 1, ception and Performance, vol. 21, no. 1, pp. 32–53, 1995. pp. S103–S109, 2001. [33] D. Huh and T. J. Sejnowski, “Spectrum of power laws for [49] G. Rizzolatti and C. Sinigaglia, “The functional role of the curved hand movements,” Proceedings of the National Acad- parieto-frontal mirror circuit: interpretations and misinterpre- emy of Sciences of the United States of America, vol. 112, tations,” Nature Reviews Neuroscience, vol. 11, no. 4, pp. 264– no. 29, pp. E3950–E3958, 2015. 274, 2010. [34] M. Zago, F. Lacquaniti, and A. Gomez-Marin, “The speed– [50] B. Calvo-Merino, J. Grèzes, D. E. Glaser, R. E. Passingham, and curvature power law in Drosophila larval locomotion,” Biology P. Haggard, “Seeing or doing? Influence of visual and motor Letters, vol. 12, no. 10, 2016. familiarity in action observation,” Current Biology, vol. 16, [35] S. Vieilledent, Y. Kerlirzin, S. Dalbera, and A. Berthoz, no. 19, pp. 1905–1910, 2006. “Relationship between velocity and curvature of a human [51] E. S. Cross, D. J. M. Kraemer, A. F. . C. Hamilton, W. M. locomotor trajectory,” Neuroscience Letters, vol. 305, no. 1, Kelley, and S. T. Grafton, “Sensitivity of the action observa- pp. 65–69, 2001. tion network to physical and observational learning,” Cere- [36] C. de’Sperati and P. Viviani, “The relationship between curva- bral Cortex, vol. 19, no. 2, pp. 315–326, 2009. ture and velocity in two-dimensional smooth pursuit eye [52] A. Neal and J. M. Kilner, “What is simulated in the action movements,” Journal of Neuroscience, vol. 17, no. 10, observation network when we observe actions?,” The Euro- pp. 3932–3945, 1997. pean Journal of Neuroscience, vol. 32, no. 10, pp. 1765–1770, [37] R. S. Marken and D. M. Shaffer, “The power law of movement: an example of a behavioral illusion,” Experimental Brain [53] M. B. Schippers and C. Keysers, “Mapping the flow of informa- Research, vol. 235, no. 6, pp. 1835–1842, 2017. tion within the putative mirror neuron system during gesture [38] R. S. Marken and D. M. Shaffer, “The power law as behavioral observation,” NeuroImage, vol. 57, no. 1, pp. 37–44, 2011. illusion: reappraising the reappraisals,” Experimental Brain [54] K. Curran and J. Lee, “Autonomic vehicles,” Computer and Research, vol. 236, no. 5, pp. 1537–1544, 2018. Information Science, vol. 1, no. 1, 2008. [39] S. B. Shafiei, “Intrinsic constraint applied by central nervous [55] J. V. Drapers, B. C. Jared, and M. W. Noakes, “Manipulator system during motor command creation of smooth hand performance evaluation using Fitts’ taping task,” in Proceed- movements,” Archives of Physical Medicine and Rehabilitation, ings of the 8th Topical Meeting on Robotics and Remote Sys- vol. 97, no. 10, article e49, 2016. tems, Pittsburgh, PA, USA, 1999. [40] P. Péruch and D. Mestre, “Between desktop and head immer- sion: functional visual field during vehicle control and naviga- tion in virtual environments,” Presence: Teleoperators and Virtual Environments, vol. 8, no. 1, pp. 54–64, 1999. [41] E. Todorov and M. I. Jordan, “Smoothness maximization along a predefined path accurately predicts the speed profiles of complex arm movements,” Journal of Neurophysiology, vol. 80, no. 2, pp. 696–714, 1998. [42] S. Lebedev, W. H. Tsui, and P. Van Gelder, “Drawing move- ments as an outcome of the principle of least action,” Journal of Mathematical Psychology, vol. 45, no. 1, pp. 43–52, 2001. [43] K. A. Guru, S. B. Shafiei, A. Khan, A. A. Hussein, M. Sharif, and E. T. Esfahani, “Understanding cognitive performance during robot-assisted surgery,” Urology, vol. 86, no. 4, pp. 751–757, 2015. [44] P. Viviani and N. Stucchi, “Biological movements look uni- form: evidence of motor-perceptual interactions,” Journal of Experimental Psychology: Human Perception and Performance, vol. 18, no. 3, pp. 603–623, 1992. [45] A. E. Patla, S. D. Prentice, C. Robinson, and J. Neufeld, “Visual control of locomotion: strategies for changing direction and for going over obstacles,” Journal of Experimental Psychology: Human Perception and Performance, vol. 17, no. 3, pp. 603– 634, 1991. International Journal of Advances in Rotating Machinery Multimedia Journal of The Scientific Journal of Engineering World Journal Sensors Hindawi Hindawi Publishing Corporation Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Journal of Control Science and Engineering Advances in Civil Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com Journal of Journal of Electrical and Computer Robotics Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 VLSI Design Advances in OptoElectronics International Journal of Modelling & Aerospace International Journal of Simulation Navigation and in Engineering Engineering Observation Hindawi Hindawi Hindawi Hindawi Volume 2018 Volume 2018 Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com www.hindawi.com www.hindawi.com Volume 2018 International Journal of Active and Passive International Journal of Antennas and Advances in Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

Journal

Applied Bionics and BiomechanicsHindawi Publishing Corporation

Published: Sep 11, 2019

References