Neural-network-enhanced torque estimation control of a soft wearable exoskeleton for elbow assistance

Liu

Proceedings of the Institution of Mechanical Engineers, Part I:

et al. 2021

This article proposes a new adaptive sliding mode repetitive learning control strategy. The proposed controller can obtain satisfactory position tracking performance in the presence of unknown dynamics and external disturbance. The unknown dynamics parameters of the exoskeleton system can be estimated via an adaptive algorithm, which is used to design the sliding mode control law. Besides, the periodic external disturbance of the system can be compensated by repetitive learning to reduce the tracking error. The stability of the proposed method is demonstrated rigorous by the Lyapunov theory. Using an upper-limb exoskeleton model, simulation results demonstrate the effectiveness of the control strategy. The proposed method has a better control performance than other methods.

Section: Simulation Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Adaptive sliding mode repetitive learning control of the upper-limb exoskeleton with unknown dynamics

Liu

Proceedings of the Institution of Mechanical Engineers, Part I:

et al. 2021

“…6 Moreover, large structure, heavy weight, high-energy consumption, and poor manmachine coupling are typical unneglectable shortcomings since the rigid robots add more extra mass and increase metabolic effort in rehabilitation training. 7 Recently, to address the above drawbacks of rigid exoskeletons, more and more researchers focus on developing the soft exoskeletons with small size, light weight, low inertia, and comfortable interaction. The Harvard University has proposed the concept of soft exoskeleton, which uses the bones and joints of the human body to replace the rigid elements of traditional exoskeleton and is combined with flexible actuators for joint actuation.…”

Section: Introductionmentioning

confidence: 99%

Design, control, and experimental verification of a soft knee exoskeleton for rehabilitation during walking

Proceedings of the Institution of Mechanical Engineers, Part I:

Chen

2021

Self Cite

Since the rigid constructions of the traditional exoskeletons are liable to increase inertia of the joint and limit wearer flexibility, the soft-type exoskeletons have been developed in recent years, which can reduce system inertia, increase interaction compliance, and ensure operation safety. In this article, we present the detailed design and preliminary control of a soft body-worn and tendon-sheath-driven exoskeleton that can apply appropriate torques to the biological knee joints during rehabilitation training. This article focuses on the design of compliance tendon-sheath actuation system based on Hill muscle model, which utilizes multi-wrap pulleys and Hill-based series elastic actuator to transmit effective forces. Besides, the overall control system and strategy architectures of knee soft exoskeleton are also proposed. A fuzzy proportional–integral–derivative controller is developed for the passive rehabilitation training. On this basis, a back-propagation-neural-network-based adaptive impedance control scheme is presented to provide adaptive force based on the disease condition of the patient. Finally, preliminary experiments are conducted to demonstrate the effectiveness of proposed exoskeleton and control strategies.

“…The assistance provided by these human joint force amplifiers can augment, reinforce or even restore human performance. Muscle strength augmentation for workers or soldiers [1], assistance to elderly generation in ADLs [2], muscle performance restoration in paraplegic patients [3], are few of the many applications of these devices. A review of existing devices and research approaches suggests that the exosuit design depends on some crucial developmental aspects: the user intention estimation, assistance torque estimation, and the comfort of force transmission [4][5][6][7].…”

Section: Introductionmentioning

confidence: 99%

An Integrated Dynamic Simulation Platform for Assistive Human-Robot Interaction: Application to Upper Limb Exosuit

Sambhav¹,

Jena²,

Chatterjee³

et al. 2021

Preprint

Soft exosuits are wearable robotic devices that assist or enhance the human muscle performance. A human machine interface simulation platform based on MATLAB-OpenSim interface is developed in this paper for closed loop dynamic simulation with feedback control strategy and to study its effect on human physiology. The proposed simulation model is based on Computed Muscle Control (CMC) algorithm and is implemented using the MATLAB -OpenSim interface. A Gravity Compensation (GC) controller has been implemented on the external device and the resulting decrease in the physiological torques, muscle activations and metabolic costs during a simple load lifting task with two different speeds is investigated.