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

Learn More →

Mechanical Design and Control Strategy for Hip Joint Power Assisting

Mechanical Design and Control Strategy for Hip Joint Power Assisting Hindawi Journal of Healthcare Engineering Volume 2018, Article ID 9712926, 7 pages https://doi.org/10.1155/2018/9712926 Research Article Mechanical Design and Control Strategy for Hip Joint Power Assisting 1,2 Wenyuan Liang College of Engineering, Peking University, Beijing 100870, China National Research Center for Rehabilitation Technical Aids, Beijing 100176, China Correspondence should be addressed to Wenyuan Liang; lwy123@hotmail.com Received 12 December 2017; Accepted 6 June 2018; Published 15 August 2018 Academic Editor: Jesus Fontecha Copyright © 2018 Wenyuan Liang. *is 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. *e basic requirements for mechanical design and control strategy are adapting to human joint movements and building an interaction model between human and robot. In this paper, a 3-UPS parallel mechanism is adopted to realize that the in- stantaneous rotation center of the assistive system coincides with human joint movement center, and a force sensory system is used to detect human movement intention and build the modeling of control strategy based on the interactive force. *en, based on the constructed experimental platform, the feasibility of movement intention detection and power assisting are verified through the experimental results. between the user and the assistive robot. During the 1. Introduction movements, the user’s joint and muscle generate the acti- Assistive robot, also called as a powered exoskeleton robot, is vation, and then the assistive robot provides the user’s joints a special robot that aims at improving the capability and with supplemental torques. In the human-machine in- efficiency of users. *e assistive robots have been developed teraction process, obtaining the human movement intention to be used for human limb strength [1], neurorehabilitation is fundamental for controlling the assistive robot. *e [2], or movement assistance [3]. *e challenges for current sensory systems are used to detect users’ movement in- research of assistive robot include the followings: how to tention or muscular activities during the process of human- design an assistive robot to adapt human movement, how to machine interaction, which could be detected directly by obtain human movement intention to promote the human- measuring EMG, interaction force sensors, gait information, machine interaction, and how to control the robot to provide and/or even EEG. HAL is developed to aid people moving an effective assistance. and executing daily-life activities. In [6], HAL mainly supports the persons with motion difficulties but whose *e design of an assistive robot should consider the physical structure of the joint and muscle to support the body EMG signals can still be detected from the muscle. NAEIES weight during movements. Hip joint, as one important part of [5] is developed for helping the user carry heavy loads, the human lower limb, is considered as a spherical joint with where the human-machine interaction force is measured by three DoFs. Some studies have designed kinds of assistive multiaxis force/torque sensors. AAFO, developed by robots for hip joint power assisting. For example, BLEEX [4] Yonsei University, uses four force sensors to detect the gait and NAEIES [5], with an anthropomorphic design, are active events [7]. on the hip flexion/extension (f/e) and abduction/adduction Based on the movement intention detection, the assistive (a/a) and passive on the hip intra/extrarotation. controllers would determine the supplemental torque pro- Movement assistance process is also called as human- vided by the assistive robot. *e controllers should ensure machine interaction, which requires a strong synergy that the provided assistive torques are coherent with human 2 Journal of Healthcare Engineering joint and muscle own activation. In [8], the robot is activated physical model can be simplified as a 3-UPS/1-S model. In with an EMG-based controller, where the desired trajectories this model, the mechanical instantaneous center of rotation in the actuators are related to the processed EMG signals of is the hip joint center. the selected muscles. In [9, 10], the actuators are driven based In this way, a 3-UPS parallel mechanism can realize 3- on the walking gait information. Model-based control [11–13] DoFs motions without disturbing the human hip joint needs to build the kinematic or dynamic model for the movement. human-machine interaction model. Since it is difficult to obtain the inertia parameters of the human limb, some model 2.1.2. Assistive Robot Modeling. As shown in Figure 3, parameters need to be estimated. assistive mechanism drives the thigh bandage (which are *e interaction processes are various among different composed of points B , B , and B ) to provide assistance for the 1 2 3 users. *e signal processing and interaction modeling are human thigh. Hence, the bandage is considered as the end- time-consuming, especially for the methods based on EMG effector of the assistive robot. *e end-effector has three DoFs, [14], EEG [15, 16], and gait information. Compared to the where parameters (Z , α _ , and β) (Figure 3) can describe end- abovementioned methods, the method based on the force effector’s movements. Z and α _ can describe end-effector’s sensor is a better choice, since different objects difference motions of abduction/adduction and flexion/extension, re- will have less influence on the model of interactive force spectively; thigh’s rotation velocity along the longitudinal axis is between human body and assistive robot. Additionally, defined as β. aiming to provide power assisting for the users who do not By considering the installation positions of motors, θ , lose muscle strength completely, the interactive force sen- θ , and θ are selected as active joints. *ese three active 22 32 sory system is considered as the intention detection method joints are driven with the brushless DC motors (Maxon EC- in this paper. 45 flat series, Figure 1). *e Jacobian representing the ve- *ough the characteristics of the joints and limbs differ locity relation between end-effector and active joints is given significantly in different users, the control process is usually as follows [17]: based on the sensory systems, which are fundamental for the T T _ _ _ _ _ control strategy. *e key of the assistive robot is to respond (1) Z α _ β � J · θ θ θ , 􏽨 􏽩 􏽨 􏽩 1 11 22 32 to human movement almost without any delay. In order to control the assistive robot with good performance on pro- where [∗] is the transpose of the matrix [∗]. Since thigh viding power assisting, the controller should be good at muscle’s movement intention would finally act on end-ef- _ _ dealing with the human-machine interaction. Considering fector’s movement, can represent thigh muscle’s 􏽨 Z α _ β 􏽩 the interaction is measured based on the force sensor, movement intention indirectly. Here, Jacobian J describes a force-based compliance controller is proposed in this the velocity relation between thigh’s movement intention paper. and actuators. *rough (1), we can correspond muscle’s *is paper is organized as follows: the second section will movement intention with an active actuator. show the hip joint assistive robot structural design and In this case, according to the force feedback between kinematic model, and the assistive robot control based on assisted limb and assistive robot, the expected end-effector the human-machine interactive force is designed; in the velocity of the human thigh with the controller can be third section, the principle for using the force sensor to generated. And then with the inverse compute based on (1), detect human movement intention is discussed, and the the expected velocities of actuators can be obtained. assistive results based on the compliance control is included; and the last section is the conclusion. 2.2. Controller Design. *e assistive mechanism is a system that provides power assisting through human-machine in- 2. Materials and Methods teraction. In the process of interaction, the key is to obtain human movement intention. *en further, in order to 2.1. Structure Design and Modeling provide power assist, it is needed to develop a control 2.1.1. Mechanical Structure Design. *e structure design for strategy based on the movement intention. the hip joint assistive robot should consider the following requirements: 2.2.1. Movement Intention Detection Based on Interactive (1) Hip joint is considered as a spherical joint of 3 DoFs, Force Sensor. In this paper, two one-dimensional force which are f/e DoF, a/a DoF, and intra/extrarotation sensors (Figure 4) are used to detect human movement DoF. intention on the motions of f/e and a/a. *e force sensor shown in Figure 4 is of high sensitivity. *ereby, it can react (2) *e assistive robot movement can cover all the three quickly to the human-machine interaction. *e detection DoFs. force f is in the sagittal plane and mainly used to detect the (3) *e assistive robot can kinematically adapt to the movement intention of extension/flexion. *e detection movement of the hip joint. force f is in the coronal plane and mainly used to detect the *en, a 3-UPS parallel mechanical assistive robot can movement intention of adduction/abduction. In this paper, meet the above requirements. As shown in Figures 1 and 2, we mainly focus on providing assistance for the movement while the human wears the hip joint assistive robot, the of extension/flexion and adduction/abduction. Journal of Healthcare Engineering 3 Actuators Force sensors (a) (b) FIGURE 1: (a) CAD model; (b) hip joint power assisting robot. A A 3 1 Hip joint center Instantaneous center of rotation (a) (b) Figure 2: A parallel mechanism for hip joint power assisting. 2.2.2. Compliant Control for Power Assisting. By online where X represents the variables in the Cartesian space, X estimating and planning the assistive torque, the proposed represents the reference commanded position, and F rep- compliance controller (shown in Figure 5) is aiming to resents the expected interaction force. In this paper, the follow human movement intention and transfer the desired expected interaction force is equal to the actual interaction assistive torque to the user’s leg eectively. force. e controller is motivated below by considering the During the assistive process, the assistive robot is ex- traditional force control model [18]: pected to follow human movement almost without any delay. at is to say, the assistive robot should not lead front x_(s) 1 R(s) , (2) or fall behind the human current position too much. f(s) M · s + B + D /s a a a erefore, the reference commanded position is given as the human current position, X  X. en, (3) is rewritten as where (2) represents the interactive force control model with d follows: the Laplace transform. M , B , and D are the inertia, a a a € _ M · X + B · X  F. (4) damping, and spring coe‚cients, respectively. a a While (2) is written as the time-domain form, it may Equation (4) also has another meaning: in (3), when the have three kinds of expression. In this paper, we consider the spring factor (D ) in the interaction model is smaller, the reference commanded position is given and unchanged. compliance eect is better; hence, when the spring factor is en, we will have too small to ignore, we can set D  0, and then we can also € _ (3) M · X + B · X + D ·  X − X  F, a a a d have the same expression as (4). Under the model shown in 4 Journal of Healthcare Engineering Waist Z γ 2 θ 3 21 Z Y θ θ 31 11 32 A Hip joint igh (X ,Y ,Z ) 2 2 2 End-effector (X ,Y ,Z ) 1 1 1 B B 3 2 (X ,Y ,Z ) 3 3 3 B Figure 3: A simpli˜ed model for the hip joint assisting robot. Interactive force f (a) (b) Figure 4: Human movement intention detection based on the force sensor. (a) Force sensor. (b) Interactive force detection. 1 + τ 1 c M V (t) − V (t) 1 –1 Assistive a n n−1 . PID Actuator · + V (t) · F (t), (7) q K n n M ·s+B robot V d t a a − B T B a a where T is the sampling cycle. e label n represents the current sampling time, and the label n − 1 represents the last Figure 5: Control strategy based on compliance control. one sampling time. ereby, the expected commanded velocity of the (4), the controller could have a better compliance, and then assistive robot end-eector, V (t), is calculated by the fol- the assistive robot could follow human movement better. lowing expression: For (4), its Laplace transform can be written as T M V(s) 1 V (t) · F (t)+ · V (t). (8) , (5) n n n−1 M + B · T M + B · T f(s) M · s + B a a a a a a where V(s) x_(s). In the time domain, (5) is expressed as In (8), the expected commanded velocity V (t) is related to the current interactive force F (t) and the previous ve- M dV(t) 1 n · + V(t)  · F(t). (6) locity V (t). e computation process can be described as n−1 B dt B a a Figure 6. Combined with (1), we can obtain the desired joint By considering the discrete form, (6) is written as follows: velocity q_ as follows: d Journal of Healthcare Engineering 5 −1 move. e force sensor reaction should be reacted quickly in stage 3, where its time cost, denoted as T , should be in the × D range T ∈ (0,T ]. In our project, the force sensor can react D B a opposite to the pressure in 1 ms. ereby, when the con- V troller obtains the human movement intention from the + n a force sensor reaction, the assistive mechanism should act in 19 ms since it detects the movement intention. T In Figure 7, the reaction force curve is obtained by the interactive force sensor, where the curve represents the movement intention of the agonist’s muscle; the actuator Figure 6: Solution of expected commanded velocity. acting trajectory belongs to the active joint of θ . It is found that the actuator acting trajectory is little lagging behind than the reaction force curve. However, the partial enlarged −1 q_  J · V (t). (9) drawings show that, after obtaining the movement intention, d n the assistive robot can act in 5∼15 ms, which is smaller than By considering the real-time velocity feedback, q, and 19 ms. PID control, then the ˜nal torque command for each joint or In short, the interactive force sensor-based movement the actuator is τ. In Figure 5, K is the torque coe‚cient for t intention detection method adopted in our project can the actuators, and then the torque command is transferred ensure the assistive robot follows human joint movement into the current command for each motor. without any delay. 3. Results and Discussion 3.2. Assistive Robot Provides Power Assisting Based on Force Sensor. Figure 8 shows that the assistive robot provides 3.1. Feasibility of Movement Intention Detection Based on power assisting while the human joint does the active Interactive Force Sensory System. Human joint movement is movement. It can be found out that the active actuators’ composed of three stages that address the following issues: acting trajectories can follow the interactive force trajecto- (1) human brain generates the movement intention, and ries well. simultaneously, the pallium would generate the relevant In Figure 8, it consists of double meanings. First, the movement nerve signals; (2) the nerve signals would interactive force curves, which represent the joint movement transmit from the brain to the agonist’s muscle corre- intention, are smooth without too much sharp jitter. is sponding the neuron, and then the neuron will induce the characteristic means that the interactive force sensors can agonist’s muscle to activate; and (3) when the agonist’s detect human movement intention exactly and without muscle activates enough, the muscle would ˜nally bring the delay, and then the force information can be used as the joint to move. Among these three stages, the ˜rst stage is input for the control. Second, based on the interactive force happening in the brain, which may be detected by the EEG; information, the compliance controller in this paper can in the second stage, the joint is not moving, but the agonist’s control the assistive robot to follow human movement al- muscle is activated which can be detected by the EMG or most without delay. It is also found that the actuators’ force sensor; and in the last stage, the joint is moving under trajectories are smooth, which means that the interaction the activation of agonist’s muscle, and its movement tra- between human and the assistive robot is with well jectory can be detected by the encoder. e sequence of these compliance. three stages is denoted by a time label, where T is the time In short, it means that force sensors can obtain the cost from the beginning of stage 1 to the end of stage 2, T is human movement intention quickly and exactly. And then, the time cost from the beginning of stage 2 to the end of stage the assistive robot can also follow human joint movement 3, and T is the time cost from the beginning of stage 1 to the quickly without delay. In this way, these experiment results end of stage 3. ensure the assistive robot can provide power assistance for According to the current literature, the transmission of the user. nerve signals from the brain to the agonist’s muscle diers according to dierent movement types. When the move- ment is performed in response to an external stimulus, the 4. Conclusions same neuron may discharge hundreds of milliseconds before a slow and accurate movement of small amplitude or only In this paper, the mechanical design and control strategy for 60∼100 ms (T ) before a ballistic movement [19]. e a parallel hip joint assistive robot are proposed. e me- ballistic movement can be detected by EMG or force sensor. chanical design is based on a 3-UPS parallel structure. e Subsequently, the triggered movements could be executed to controller design is based on compliance control with in- act with joint movements in 80∼120 ms (T ) [20]. teractive force sensors. e experiment results show that the e time cost of the human joint ready to move, denoted interactive force-based movement intention detection is as (T  T − T ), is about 20 ms. T means the time cost available, and the compliance controller also has a good B C A B from the time node that the nerve signals induce the performance in following human movements by providing muscle activation to the time node that the joint starts to power assist. 6 Journal of Healthcare Engineering coe‚cients. (2) e compliance proposed in this paper has a good performance during the human movement process. However, as shown in the formula of (8), the controller could not have a good performance while the interactive 15 ms force equals to zero. (3) e assistive eect is needed to be 5 ms assessed via the EMG to detect the activation dierence of the agonist’s muscle with and without assisting. 10 10 Conflicts of Interest e author declares that there are no con¢icts of interest 5 5 regarding the publication of this paper. Acknowledgments 0 0 is work was supported by the Fundamental Research Funds for Central Public Welfare Research Institutes –5 –5 (118009001000160001). 15 ms 5ms 4000 4100 4200 4300 4400 4500 4600 4700 4800 4000 5000 References Time (ms) [1] H. Kazerooni, “Human-robot interaction via the transfer of Interactive force power and information signals,” IEEE Transactions on Sys- Active joint velocity tems, Man, and Cybernetics, vol. 20, no. 2, pp. 450–463, 1990. Figure 7: Human movement intention based on interactive force [2] S. K. Banala, S. H. Kim, S. K. Agrawal, and J. P. Scholz, “Robot sensor reaction. assisted gait training with active leg exoskeleton (ALEX),” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 17, no. 1, pp. 2–8, 2009. 24 12 [3] R. J. Farris, H. Quintero, and M. Goldfarb, “Preliminary Interactive force evaluation of a powered lower limb orthosis to aid walking in paraplegic individuals,” IEEE Transactions on Neural Systems 14 θ 7 and Rehabilitation Engineering, vol. 19, no. 6, pp. 652–659, 4 2 N [4] H. Kazerooni, “Exoskeletons for human power augmenta- tion,” in Proceedings of 2005 IEEE/RSJ International Con- ference on Intelligent Robots and System, pp. 3120–3125, -6 –3 Alberta, Canada, 2005. [5] Z. Yang, Y. Zhu, X. Yang, and Y. Zhang, “Impedance control –16 –8 of exoskeleton suit based on adaptive RBF neural network,” in Proceedings of International Conference on Intelligent Human Time (ms) Machine Systems and Cybernetics, pp. 182–187, Zhejiang, China, August 2009. Interactive force [6] H. Kawamoto, S. Taal, H. Niniss et al., “Voluntary motion support control of robot suit HAL triggered by bioelectrical signal for hemiplegia,” in Proceedings of 2010 Annual In- ternational Conference of the IEEE Engineering in Medicine and Biology Society, pp. 462–466, Buenos Aires, Argentina, August 2010. [7] J. Kim, S. Hwang, and Y. Kim, “Development of an active ankle- foot orthosis for hemiplegic patients,” in Proceedings of the 1st –5 –5 International Convention on Rehabilitation Engineering & Assistive Technology in Conjunction with 1st Tan Tock Seng Time (ms) Hospital Neurorehabilitation Meeting- i-CREATe’07, pp.110–113, New York, NY, USA, 2007. Figure 8: Force sensor interactive trajectories and actuators’ [8] C. Fleischer and G. Hommel, “A human exoskeleton interface tracking trajectories. utilizing electromyography,” IEEE Transactions on Robotics, vol. 24, no. 4, pp. 872–882, 2008. [9] D. Sasaki, T. Noritsugu, and M. Takaiwa, “Development of In the future work, it needs to address three issues: (1) pneumatic lower limb power assist wear driven with wearable in this paper, the coe‚cients of M and B are determined air supply system,” in Proceedings of 2013 IEEE/RSJ In- a a by many times of trials. We would like to use much ternational Conference on Proceedings of Intelligent Robots adaptive optimized method to determine these two and Systems, IROS, pp. 4440–4445, November 2013. Rad/s Rad/s Rad/s Journal of Healthcare Engineering 7 [10] A. T. Asbeck, R. J. Dyer, A. F. Larusson, and C. J. Walsh, “Biologically-inspired soft exosuit,” in Proceedings of IEEE. International Conference on Rehabilitation Robotics, p. 6650455, June 2013. [11] Y. Yu, W. Liang, and Y. Ge, “Jacobian analysis for parallel mechanism using on human walking power assisting,” in Proceedings of 2011 International Conference on Mechatronics and Automation, ICMA, pp. 282–288, August 2011. [12] G. Aguirre-Ollinger, J. E. Colgate, M. A. Peshkin, and A. Goswami, “Inertia compensation control of a one-degree- of-freedom exoskeleton for lower-limb assistance: initial ex- periments,” IEEE Transactions on Neural Systems and Re- habilitation Engineering, vol. 20, no. 1, pp. 68–77, 2012. [13] A. Polinkovsky, R. J. Bachmann, N. I. Kern, and R. D. Quinn, “An ankle foot orthosis with insertion point eccentricity control,” in Proceedings of 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS, pp. 1603–1608, Vilamoura, Portugal, 2012. [14] Y. Yu and W. Y. Liang, “Manipulability inclusive principle for hip joint assistive mechanism design optimization,” In- ternational Journal of Advanced Manufacturing Technology, vol. 70, no. 5–8, pp. 929–945, 2014. [15] N. Hogan, “Impedance control: an approach to manipulation, part I—theory,” Journal of Dynamic Systems, Measurement, and Control, vol. 107, no. 1, pp. 1–7, 1985. [16] L. Leocani, L. G. Cohen, E. M. Wassermann, K. Ikoma, and M. Hallett, “Human corticospinal excitability evaluated with transcranial magnetic stimulation during different reaction time paradigms,” Journal of Brain, vol. 123, no. 6, pp. 1161– 1173, 2000. [17] L. M. Nashner and P. J. Cordo, “Relation of automatic postural responses and reaction-time voluntary movement of human leg muscles,” Jounal of Experimental Brain Reserach, vol. 43, no. 3-4, pp. 395–406, 1981. [18] B. Chen, L. Grazi, F. Lanotte, N. Vitiello, and S. Crea, “A real- time lift detection strategy for a hip exoskeleton,” Frontiers in Neurorobotics, vol. 12, p. 17, 2018. [19] A. Kostov and M. Polak, “Parallel man-machine training in development of EEG-based cursor control,” IEEE Trans- actions on Rehabilitation Engineering, vol. 8, no. 2, pp. 203– 205, 2000. [20] D. Wu, V. J. Lawhern, and B. J. Lance, “Reducing offline BCI calibration effort using weighted adaptation regularization with source domain selection,” in Proceedings of IEEE In- ternational Conference on Systems, Man, and Cybernetics, pp. 3209–3216, Budapest, Hungary, 2016. 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 Journal of Healthcare Engineering Hindawi Publishing Corporation

Mechanical Design and Control Strategy for Hip Joint Power Assisting

Liang, Wenyuan
Journal of Healthcare Engineering , Volume 2018: 7 – Aug 15, 2018
Download PDF
8 pages

Loading next page...
 
/lp/hindawi-publishing-corporation/mechanical-design-and-control-strategy-for-hip-joint-power-assisting-2t6Vmo8MP0

References (27)

Publisher
Hindawi Publishing Corporation
Copyright
Copyright © 2018 Wenyuan Liang. 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
2040-2295
eISSN
2040-2309
DOI
10.1155/2018/9712926
Publisher site
See Article on Publisher Site

Abstract

Hindawi Journal of Healthcare Engineering Volume 2018, Article ID 9712926, 7 pages https://doi.org/10.1155/2018/9712926 Research Article Mechanical Design and Control Strategy for Hip Joint Power Assisting 1,2 Wenyuan Liang College of Engineering, Peking University, Beijing 100870, China National Research Center for Rehabilitation Technical Aids, Beijing 100176, China Correspondence should be addressed to Wenyuan Liang; lwy123@hotmail.com Received 12 December 2017; Accepted 6 June 2018; Published 15 August 2018 Academic Editor: Jesus Fontecha Copyright © 2018 Wenyuan Liang. *is 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. *e basic requirements for mechanical design and control strategy are adapting to human joint movements and building an interaction model between human and robot. In this paper, a 3-UPS parallel mechanism is adopted to realize that the in- stantaneous rotation center of the assistive system coincides with human joint movement center, and a force sensory system is used to detect human movement intention and build the modeling of control strategy based on the interactive force. *en, based on the constructed experimental platform, the feasibility of movement intention detection and power assisting are verified through the experimental results. between the user and the assistive robot. During the 1. Introduction movements, the user’s joint and muscle generate the acti- Assistive robot, also called as a powered exoskeleton robot, is vation, and then the assistive robot provides the user’s joints a special robot that aims at improving the capability and with supplemental torques. In the human-machine in- efficiency of users. *e assistive robots have been developed teraction process, obtaining the human movement intention to be used for human limb strength [1], neurorehabilitation is fundamental for controlling the assistive robot. *e [2], or movement assistance [3]. *e challenges for current sensory systems are used to detect users’ movement in- research of assistive robot include the followings: how to tention or muscular activities during the process of human- design an assistive robot to adapt human movement, how to machine interaction, which could be detected directly by obtain human movement intention to promote the human- measuring EMG, interaction force sensors, gait information, machine interaction, and how to control the robot to provide and/or even EEG. HAL is developed to aid people moving an effective assistance. and executing daily-life activities. In [6], HAL mainly supports the persons with motion difficulties but whose *e design of an assistive robot should consider the physical structure of the joint and muscle to support the body EMG signals can still be detected from the muscle. NAEIES weight during movements. Hip joint, as one important part of [5] is developed for helping the user carry heavy loads, the human lower limb, is considered as a spherical joint with where the human-machine interaction force is measured by three DoFs. Some studies have designed kinds of assistive multiaxis force/torque sensors. AAFO, developed by robots for hip joint power assisting. For example, BLEEX [4] Yonsei University, uses four force sensors to detect the gait and NAEIES [5], with an anthropomorphic design, are active events [7]. on the hip flexion/extension (f/e) and abduction/adduction Based on the movement intention detection, the assistive (a/a) and passive on the hip intra/extrarotation. controllers would determine the supplemental torque pro- Movement assistance process is also called as human- vided by the assistive robot. *e controllers should ensure machine interaction, which requires a strong synergy that the provided assistive torques are coherent with human 2 Journal of Healthcare Engineering joint and muscle own activation. In [8], the robot is activated physical model can be simplified as a 3-UPS/1-S model. In with an EMG-based controller, where the desired trajectories this model, the mechanical instantaneous center of rotation in the actuators are related to the processed EMG signals of is the hip joint center. the selected muscles. In [9, 10], the actuators are driven based In this way, a 3-UPS parallel mechanism can realize 3- on the walking gait information. Model-based control [11–13] DoFs motions without disturbing the human hip joint needs to build the kinematic or dynamic model for the movement. human-machine interaction model. Since it is difficult to obtain the inertia parameters of the human limb, some model 2.1.2. Assistive Robot Modeling. As shown in Figure 3, parameters need to be estimated. assistive mechanism drives the thigh bandage (which are *e interaction processes are various among different composed of points B , B , and B ) to provide assistance for the 1 2 3 users. *e signal processing and interaction modeling are human thigh. Hence, the bandage is considered as the end- time-consuming, especially for the methods based on EMG effector of the assistive robot. *e end-effector has three DoFs, [14], EEG [15, 16], and gait information. Compared to the where parameters (Z , α _ , and β) (Figure 3) can describe end- abovementioned methods, the method based on the force effector’s movements. Z and α _ can describe end-effector’s sensor is a better choice, since different objects difference motions of abduction/adduction and flexion/extension, re- will have less influence on the model of interactive force spectively; thigh’s rotation velocity along the longitudinal axis is between human body and assistive robot. Additionally, defined as β. aiming to provide power assisting for the users who do not By considering the installation positions of motors, θ , lose muscle strength completely, the interactive force sen- θ , and θ are selected as active joints. *ese three active 22 32 sory system is considered as the intention detection method joints are driven with the brushless DC motors (Maxon EC- in this paper. 45 flat series, Figure 1). *e Jacobian representing the ve- *ough the characteristics of the joints and limbs differ locity relation between end-effector and active joints is given significantly in different users, the control process is usually as follows [17]: based on the sensory systems, which are fundamental for the T T _ _ _ _ _ control strategy. *e key of the assistive robot is to respond (1) Z α _ β � J · θ θ θ , 􏽨 􏽩 􏽨 􏽩 1 11 22 32 to human movement almost without any delay. In order to control the assistive robot with good performance on pro- where [∗] is the transpose of the matrix [∗]. Since thigh viding power assisting, the controller should be good at muscle’s movement intention would finally act on end-ef- _ _ dealing with the human-machine interaction. Considering fector’s movement, can represent thigh muscle’s 􏽨 Z α _ β 􏽩 the interaction is measured based on the force sensor, movement intention indirectly. Here, Jacobian J describes a force-based compliance controller is proposed in this the velocity relation between thigh’s movement intention paper. and actuators. *rough (1), we can correspond muscle’s *is paper is organized as follows: the second section will movement intention with an active actuator. show the hip joint assistive robot structural design and In this case, according to the force feedback between kinematic model, and the assistive robot control based on assisted limb and assistive robot, the expected end-effector the human-machine interactive force is designed; in the velocity of the human thigh with the controller can be third section, the principle for using the force sensor to generated. And then with the inverse compute based on (1), detect human movement intention is discussed, and the the expected velocities of actuators can be obtained. assistive results based on the compliance control is included; and the last section is the conclusion. 2.2. Controller Design. *e assistive mechanism is a system that provides power assisting through human-machine in- 2. Materials and Methods teraction. In the process of interaction, the key is to obtain human movement intention. *en further, in order to 2.1. Structure Design and Modeling provide power assist, it is needed to develop a control 2.1.1. Mechanical Structure Design. *e structure design for strategy based on the movement intention. the hip joint assistive robot should consider the following requirements: 2.2.1. Movement Intention Detection Based on Interactive (1) Hip joint is considered as a spherical joint of 3 DoFs, Force Sensor. In this paper, two one-dimensional force which are f/e DoF, a/a DoF, and intra/extrarotation sensors (Figure 4) are used to detect human movement DoF. intention on the motions of f/e and a/a. *e force sensor shown in Figure 4 is of high sensitivity. *ereby, it can react (2) *e assistive robot movement can cover all the three quickly to the human-machine interaction. *e detection DoFs. force f is in the sagittal plane and mainly used to detect the (3) *e assistive robot can kinematically adapt to the movement intention of extension/flexion. *e detection movement of the hip joint. force f is in the coronal plane and mainly used to detect the *en, a 3-UPS parallel mechanical assistive robot can movement intention of adduction/abduction. In this paper, meet the above requirements. As shown in Figures 1 and 2, we mainly focus on providing assistance for the movement while the human wears the hip joint assistive robot, the of extension/flexion and adduction/abduction. Journal of Healthcare Engineering 3 Actuators Force sensors (a) (b) FIGURE 1: (a) CAD model; (b) hip joint power assisting robot. A A 3 1 Hip joint center Instantaneous center of rotation (a) (b) Figure 2: A parallel mechanism for hip joint power assisting. 2.2.2. Compliant Control for Power Assisting. By online where X represents the variables in the Cartesian space, X estimating and planning the assistive torque, the proposed represents the reference commanded position, and F rep- compliance controller (shown in Figure 5) is aiming to resents the expected interaction force. In this paper, the follow human movement intention and transfer the desired expected interaction force is equal to the actual interaction assistive torque to the user’s leg eectively. force. e controller is motivated below by considering the During the assistive process, the assistive robot is ex- traditional force control model [18]: pected to follow human movement almost without any delay. at is to say, the assistive robot should not lead front x_(s) 1 R(s) , (2) or fall behind the human current position too much. f(s) M · s + B + D /s a a a erefore, the reference commanded position is given as the human current position, X  X. en, (3) is rewritten as where (2) represents the interactive force control model with d follows: the Laplace transform. M , B , and D are the inertia, a a a € _ M · X + B · X  F. (4) damping, and spring coe‚cients, respectively. a a While (2) is written as the time-domain form, it may Equation (4) also has another meaning: in (3), when the have three kinds of expression. In this paper, we consider the spring factor (D ) in the interaction model is smaller, the reference commanded position is given and unchanged. compliance eect is better; hence, when the spring factor is en, we will have too small to ignore, we can set D  0, and then we can also € _ (3) M · X + B · X + D ·  X − X  F, a a a d have the same expression as (4). Under the model shown in 4 Journal of Healthcare Engineering Waist Z γ 2 θ 3 21 Z Y θ θ 31 11 32 A Hip joint igh (X ,Y ,Z ) 2 2 2 End-effector (X ,Y ,Z ) 1 1 1 B B 3 2 (X ,Y ,Z ) 3 3 3 B Figure 3: A simpli˜ed model for the hip joint assisting robot. Interactive force f (a) (b) Figure 4: Human movement intention detection based on the force sensor. (a) Force sensor. (b) Interactive force detection. 1 + τ 1 c M V (t) − V (t) 1 –1 Assistive a n n−1 . PID Actuator · + V (t) · F (t), (7) q K n n M ·s+B robot V d t a a − B T B a a where T is the sampling cycle. e label n represents the current sampling time, and the label n − 1 represents the last Figure 5: Control strategy based on compliance control. one sampling time. ereby, the expected commanded velocity of the (4), the controller could have a better compliance, and then assistive robot end-eector, V (t), is calculated by the fol- the assistive robot could follow human movement better. lowing expression: For (4), its Laplace transform can be written as T M V(s) 1 V (t) · F (t)+ · V (t). (8) , (5) n n n−1 M + B · T M + B · T f(s) M · s + B a a a a a a where V(s) x_(s). In the time domain, (5) is expressed as In (8), the expected commanded velocity V (t) is related to the current interactive force F (t) and the previous ve- M dV(t) 1 n · + V(t)  · F(t). (6) locity V (t). e computation process can be described as n−1 B dt B a a Figure 6. Combined with (1), we can obtain the desired joint By considering the discrete form, (6) is written as follows: velocity q_ as follows: d Journal of Healthcare Engineering 5 −1 move. e force sensor reaction should be reacted quickly in stage 3, where its time cost, denoted as T , should be in the × D range T ∈ (0,T ]. In our project, the force sensor can react D B a opposite to the pressure in 1 ms. ereby, when the con- V troller obtains the human movement intention from the + n a force sensor reaction, the assistive mechanism should act in 19 ms since it detects the movement intention. T In Figure 7, the reaction force curve is obtained by the interactive force sensor, where the curve represents the movement intention of the agonist’s muscle; the actuator Figure 6: Solution of expected commanded velocity. acting trajectory belongs to the active joint of θ . It is found that the actuator acting trajectory is little lagging behind than the reaction force curve. However, the partial enlarged −1 q_  J · V (t). (9) drawings show that, after obtaining the movement intention, d n the assistive robot can act in 5∼15 ms, which is smaller than By considering the real-time velocity feedback, q, and 19 ms. PID control, then the ˜nal torque command for each joint or In short, the interactive force sensor-based movement the actuator is τ. In Figure 5, K is the torque coe‚cient for t intention detection method adopted in our project can the actuators, and then the torque command is transferred ensure the assistive robot follows human joint movement into the current command for each motor. without any delay. 3. Results and Discussion 3.2. Assistive Robot Provides Power Assisting Based on Force Sensor. Figure 8 shows that the assistive robot provides 3.1. Feasibility of Movement Intention Detection Based on power assisting while the human joint does the active Interactive Force Sensory System. Human joint movement is movement. It can be found out that the active actuators’ composed of three stages that address the following issues: acting trajectories can follow the interactive force trajecto- (1) human brain generates the movement intention, and ries well. simultaneously, the pallium would generate the relevant In Figure 8, it consists of double meanings. First, the movement nerve signals; (2) the nerve signals would interactive force curves, which represent the joint movement transmit from the brain to the agonist’s muscle corre- intention, are smooth without too much sharp jitter. is sponding the neuron, and then the neuron will induce the characteristic means that the interactive force sensors can agonist’s muscle to activate; and (3) when the agonist’s detect human movement intention exactly and without muscle activates enough, the muscle would ˜nally bring the delay, and then the force information can be used as the joint to move. Among these three stages, the ˜rst stage is input for the control. Second, based on the interactive force happening in the brain, which may be detected by the EEG; information, the compliance controller in this paper can in the second stage, the joint is not moving, but the agonist’s control the assistive robot to follow human movement al- muscle is activated which can be detected by the EMG or most without delay. It is also found that the actuators’ force sensor; and in the last stage, the joint is moving under trajectories are smooth, which means that the interaction the activation of agonist’s muscle, and its movement tra- between human and the assistive robot is with well jectory can be detected by the encoder. e sequence of these compliance. three stages is denoted by a time label, where T is the time In short, it means that force sensors can obtain the cost from the beginning of stage 1 to the end of stage 2, T is human movement intention quickly and exactly. And then, the time cost from the beginning of stage 2 to the end of stage the assistive robot can also follow human joint movement 3, and T is the time cost from the beginning of stage 1 to the quickly without delay. In this way, these experiment results end of stage 3. ensure the assistive robot can provide power assistance for According to the current literature, the transmission of the user. nerve signals from the brain to the agonist’s muscle diers according to dierent movement types. When the move- ment is performed in response to an external stimulus, the 4. Conclusions same neuron may discharge hundreds of milliseconds before a slow and accurate movement of small amplitude or only In this paper, the mechanical design and control strategy for 60∼100 ms (T ) before a ballistic movement [19]. e a parallel hip joint assistive robot are proposed. e me- ballistic movement can be detected by EMG or force sensor. chanical design is based on a 3-UPS parallel structure. e Subsequently, the triggered movements could be executed to controller design is based on compliance control with in- act with joint movements in 80∼120 ms (T ) [20]. teractive force sensors. e experiment results show that the e time cost of the human joint ready to move, denoted interactive force-based movement intention detection is as (T  T − T ), is about 20 ms. T means the time cost available, and the compliance controller also has a good B C A B from the time node that the nerve signals induce the performance in following human movements by providing muscle activation to the time node that the joint starts to power assist. 6 Journal of Healthcare Engineering coe‚cients. (2) e compliance proposed in this paper has a good performance during the human movement process. However, as shown in the formula of (8), the controller could not have a good performance while the interactive 15 ms force equals to zero. (3) e assistive eect is needed to be 5 ms assessed via the EMG to detect the activation dierence of the agonist’s muscle with and without assisting. 10 10 Conflicts of Interest e author declares that there are no con¢icts of interest 5 5 regarding the publication of this paper. Acknowledgments 0 0 is work was supported by the Fundamental Research Funds for Central Public Welfare Research Institutes –5 –5 (118009001000160001). 15 ms 5ms 4000 4100 4200 4300 4400 4500 4600 4700 4800 4000 5000 References Time (ms) [1] H. Kazerooni, “Human-robot interaction via the transfer of Interactive force power and information signals,” IEEE Transactions on Sys- Active joint velocity tems, Man, and Cybernetics, vol. 20, no. 2, pp. 450–463, 1990. Figure 7: Human movement intention based on interactive force [2] S. K. Banala, S. H. Kim, S. K. Agrawal, and J. P. Scholz, “Robot sensor reaction. assisted gait training with active leg exoskeleton (ALEX),” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 17, no. 1, pp. 2–8, 2009. 24 12 [3] R. J. Farris, H. Quintero, and M. Goldfarb, “Preliminary Interactive force evaluation of a powered lower limb orthosis to aid walking in paraplegic individuals,” IEEE Transactions on Neural Systems 14 θ 7 and Rehabilitation Engineering, vol. 19, no. 6, pp. 652–659, 4 2 N [4] H. Kazerooni, “Exoskeletons for human power augmenta- tion,” in Proceedings of 2005 IEEE/RSJ International Con- ference on Intelligent Robots and System, pp. 3120–3125, -6 –3 Alberta, Canada, 2005. [5] Z. Yang, Y. Zhu, X. Yang, and Y. Zhang, “Impedance control –16 –8 of exoskeleton suit based on adaptive RBF neural network,” in Proceedings of International Conference on Intelligent Human Time (ms) Machine Systems and Cybernetics, pp. 182–187, Zhejiang, China, August 2009. Interactive force [6] H. Kawamoto, S. Taal, H. Niniss et al., “Voluntary motion support control of robot suit HAL triggered by bioelectrical signal for hemiplegia,” in Proceedings of 2010 Annual In- ternational Conference of the IEEE Engineering in Medicine and Biology Society, pp. 462–466, Buenos Aires, Argentina, August 2010. [7] J. Kim, S. Hwang, and Y. Kim, “Development of an active ankle- foot orthosis for hemiplegic patients,” in Proceedings of the 1st –5 –5 International Convention on Rehabilitation Engineering & Assistive Technology in Conjunction with 1st Tan Tock Seng Time (ms) Hospital Neurorehabilitation Meeting- i-CREATe’07, pp.110–113, New York, NY, USA, 2007. Figure 8: Force sensor interactive trajectories and actuators’ [8] C. Fleischer and G. Hommel, “A human exoskeleton interface tracking trajectories. utilizing electromyography,” IEEE Transactions on Robotics, vol. 24, no. 4, pp. 872–882, 2008. [9] D. Sasaki, T. Noritsugu, and M. Takaiwa, “Development of In the future work, it needs to address three issues: (1) pneumatic lower limb power assist wear driven with wearable in this paper, the coe‚cients of M and B are determined air supply system,” in Proceedings of 2013 IEEE/RSJ In- a a by many times of trials. We would like to use much ternational Conference on Proceedings of Intelligent Robots adaptive optimized method to determine these two and Systems, IROS, pp. 4440–4445, November 2013. Rad/s Rad/s Rad/s Journal of Healthcare Engineering 7 [10] A. T. Asbeck, R. J. Dyer, A. F. Larusson, and C. J. Walsh, “Biologically-inspired soft exosuit,” in Proceedings of IEEE. International Conference on Rehabilitation Robotics, p. 6650455, June 2013. [11] Y. Yu, W. Liang, and Y. Ge, “Jacobian analysis for parallel mechanism using on human walking power assisting,” in Proceedings of 2011 International Conference on Mechatronics and Automation, ICMA, pp. 282–288, August 2011. [12] G. Aguirre-Ollinger, J. E. Colgate, M. A. Peshkin, and A. Goswami, “Inertia compensation control of a one-degree- of-freedom exoskeleton for lower-limb assistance: initial ex- periments,” IEEE Transactions on Neural Systems and Re- habilitation Engineering, vol. 20, no. 1, pp. 68–77, 2012. [13] A. Polinkovsky, R. J. Bachmann, N. I. Kern, and R. D. Quinn, “An ankle foot orthosis with insertion point eccentricity control,” in Proceedings of 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS, pp. 1603–1608, Vilamoura, Portugal, 2012. [14] Y. Yu and W. Y. Liang, “Manipulability inclusive principle for hip joint assistive mechanism design optimization,” In- ternational Journal of Advanced Manufacturing Technology, vol. 70, no. 5–8, pp. 929–945, 2014. [15] N. Hogan, “Impedance control: an approach to manipulation, part I—theory,” Journal of Dynamic Systems, Measurement, and Control, vol. 107, no. 1, pp. 1–7, 1985. [16] L. Leocani, L. G. Cohen, E. M. Wassermann, K. Ikoma, and M. Hallett, “Human corticospinal excitability evaluated with transcranial magnetic stimulation during different reaction time paradigms,” Journal of Brain, vol. 123, no. 6, pp. 1161– 1173, 2000. [17] L. M. Nashner and P. J. Cordo, “Relation of automatic postural responses and reaction-time voluntary movement of human leg muscles,” Jounal of Experimental Brain Reserach, vol. 43, no. 3-4, pp. 395–406, 1981. [18] B. Chen, L. Grazi, F. Lanotte, N. Vitiello, and S. Crea, “A real- time lift detection strategy for a hip exoskeleton,” Frontiers in Neurorobotics, vol. 12, p. 17, 2018. [19] A. Kostov and M. Polak, “Parallel man-machine training in development of EEG-based cursor control,” IEEE Trans- actions on Rehabilitation Engineering, vol. 8, no. 2, pp. 203– 205, 2000. [20] D. Wu, V. J. Lawhern, and B. J. Lance, “Reducing offline BCI calibration effort using weighted adaptation regularization with source domain selection,” in Proceedings of IEEE In- ternational Conference on Systems, Man, and Cybernetics, pp. 3209–3216, Budapest, Hungary, 2016. 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

Journal of Healthcare Engineering – Hindawi Publishing Corporation

Published: Aug 15, 2018

There are no references for this article.