Robust Iterative Learning Control for Pneumatic Muscle With Uncertainties and State Constraints

Qian, Kun; Li, Zhenhong; Chakrabarty, Samit; Zhang, Zhiqiang; Xie, Sheng Quan

doi:10.1109/tie.2022.3159970

Cited by 20 publications

(4 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The compliance of pneumatic muscle could improve the system's flexibility, safety, and comfort of the human-machine system. Moreover, the muscle has advantages such as the simple structure, light weight, and high powerweight ratio, compared with conventional electric motors [24], [25]. The sensory system includes two load cells connected in line with the pneumatic muscles respectively.…”

Section: A Wrist Exoskeletonmentioning

confidence: 99%

Adaptive Cooperative Control Strategy for a Wrist Exoskeleton Using Model-Based Joint Impedance Estimation

Zhao

Qian

Sheng

et al. 2023

IEEE/ASME Trans. Mechatron.

View full text Add to dashboard Cite

Wrist rehabilitation exoskeletons have gained much attention over the last decades, striving to restore motor functions for patients with neuromuscular disorders. Electromyography (EMG) signal has been employed to estimate the motion intention to achieve interactive training schemes. However, it is a challenging task to estimate the joint impedance in realtime, as it is a crucial parameter for control of exoskeletons. This paper proposes an adaptive cooperative control strategy for a wrist exoskeleton based on a real-time joint impedance estimation approach. By explicitly interpreting the underlying transformation in the muscular and skeletal systems, the proposed approach estimates the motion intention and the joint impedance of a human subject simultaneously without additional calibration procedures and regulates the training trajectories and assistance accordingly. Results indicate the proposed method outperforms other training protocols, including the trajectory tracking control and the fixed cooperative control. The proposed control strategy provides an additional 66.25% motion deviation when estimated joint torque increases 12.36%, which enhances the training effectiveness and the interaction safety and promotes subjects' active engagement.

show abstract

Section: A Wrist Exoskeletonmentioning

confidence: 99%

Adaptive Cooperative Control Strategy for a Wrist Exoskeleton Using Model-Based Joint Impedance Estimation

Zhao

Qian

Sheng

et al. 2023

IEEE/ASME Trans. Mechatron.

View full text Add to dashboard Cite

show abstract

“…Recent research uses ILC for a variety of tasks such as friction correction, trajectory or position tracking, voltage stabilization, or to adjust control quantity. Riaz et al [9] employ ILC to correct friction at low speeds, while [10] utilized it for trajectory tracking of pneumatic muscle actuators with state constraints. Suh et al [11] propose a current-error-based ILC approach with a nonlinear controller to improve position-tracking performance in permanent magnet stepper motors.…”

Section: Iterative Learning Controlmentioning

confidence: 99%

Current Harmonics Minimization of Permanent Magnet Synchronous Machine Based on Iterative Learning Control and Neural Networks

Mai,

Liu,

Wagner

et al. 2023

Machines

View full text Add to dashboard Cite

Electrical machines generate unwanted flux and current harmonics. Harmonics can be suppressed using various methods. In this paper, the harmonics are significantly reduced using Iterative Learning Control (ILC) and Neural Networks (NNs). The ILC can compensate for the harmonics well for operation at constant speed and current reference values. The NNs are trained with the data from the ILC and help to suppress the harmonics well even in transient operation. The simulation model is based on flux and torque maps, depending on dq-currents and the electrical angle. The maps are generated from FEM simulation of an interior permanent magnet synchronous machine (IPM) and are published with the paper. They are intended to serve other researchers for direct comparison with their own methods. Simulation results in this paper verify that by using ILC and NNs together, current harmonics in transient operation can be eliminated better than without NNs.

show abstract

“…Among them, the Huxley-type model is more in line with the structure of real muscle anatomy, but it relies on a large number of prior parameters and mathematically relies on complex partial differential equations, whereas the Hill-type model is the most widely adopted muscle computational model due to its simple and direct modeling process as well as its high degree of fitting to the real muscle fundamental dynamics. In the hardware research of muscle modules, there are three paths to achieve humanoid muscle, namely, the wire-pulling muscle scheme (Yuan et al, 2023;Fan et al, 2023), the pneumatic muscle scheme (Qian et al, 2022;Hamon et al, 2023) and the scheme using new materials and technologies (Na et al, 2023;Kobayashi et al, 2023). However, in both simulated muscle models and actual hardware muscle modules, humanoid muscles are highly redundant, nonlinear and complex objects, and the modeling process involves a large number of muscle state variables, while most of the relevant variables of muscles are unknown or unmeasurable due to the types of sensors.…”

Section: Introductionmentioning

confidence: 99%

A memory and attention-based reinforcement learning for musculoskeletal robots with prior knowledge of muscle synergies

Wang,

Chen,

Qiao

2024

RIA

View full text Add to dashboard Cite

Purpose Limited by the types of sensors, the state information available for musculoskeletal robots with highly redundant, nonlinear muscles is often incomplete, which makes the control face a bottleneck problem. The aim of this paper is to design a method to improve the motion performance of musculoskeletal robots in partially observable scenarios, and to leverage the ontology knowledge to enhance the algorithm’s adaptability to musculoskeletal robots that have undergone changes. Design/methodology/approach A memory and attention-based reinforcement learning method is proposed for musculoskeletal robots with prior knowledge of muscle synergies. First, to deal with partially observed states available to musculoskeletal robots, a memory and attention-based network architecture is proposed for inferring more sufficient and intrinsic states. Second, inspired by muscle synergy hypothesis in neuroscience, prior knowledge of a musculoskeletal robot’s muscle synergies is embedded in network structure and reward shaping. Findings Based on systematic validation, it is found that the proposed method demonstrates superiority over the traditional twin delayed deep deterministic policy gradients (TD3) algorithm. A musculoskeletal robot with highly redundant, nonlinear muscles is adopted to implement goal-directed tasks. In the case of 21-dimensional states, the learning efficiency and accuracy are significantly improved compared with the traditional TD3 algorithm; in the case of 13-dimensional states without velocities and information from the end effector, the traditional TD3 is unable to complete the reaching tasks, while the proposed method breaks through this bottleneck problem. Originality/value In this paper, a novel memory and attention-based reinforcement learning method with prior knowledge of muscle synergies is proposed for musculoskeletal robots to deal with partially observable scenarios. Compared with the existing methods, the proposed method effectively improves the performance. Furthermore, this paper promotes the fusion of neuroscience and robotics.

show abstract

Robust Iterative Learning Control for Pneumatic Muscle With Uncertainties and State Constraints

Abstract: This is a repository copy of Robust Iterative Learning Control for Pneumatic Muscle with Uncertainties and State Constraints.

Cited by 20 publications

References 31 publications

Adaptive Cooperative Control Strategy for a Wrist Exoskeleton Using Model-Based Joint Impedance Estimation

Adaptive Cooperative Control Strategy for a Wrist Exoskeleton Using Model-Based Joint Impedance Estimation

Current Harmonics Minimization of Permanent Magnet Synchronous Machine Based on Iterative Learning Control and Neural Networks

A memory and attention-based reinforcement learning for musculoskeletal robots with prior knowledge of muscle synergies

Contact Info

Product

Resources

About