Neural network based iterative learning control for magnetic shape memory alloy actuator with iteration-dependent uncertainties

Yu, Yewei; Zhang, Chen; Cao, Wenjing; Zhang, Xiuyu; Zhou, Miaolei

doi:10.1016/j.ymssp.2022.109950

Cited by 31 publications

(5 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some research groups have explored the use of a combination of non-parametric and parametric models to describe the asymmetric hysteresis behaviour observed in some cases [27,28]. MR clutches exhibit symmetric hysteresis behaviour for the most part; as such, it is not necessary to resort to such hybrid approaches in order to avoid the extra complexity of the model.…”

Section: Overview Of Hysteresis Modeling In Mr Clutchesmentioning

confidence: 99%

A Computationally Efficient Hysteresis Model for Magneto-Rheological Clutches and Its Comparison with Other Models

Yang

Kermani

2023

Actuators

View full text Add to dashboard Cite

The collaborative robot market has experienced rapid growth, leading to advancements in compliant actuation and torque control. Magneto-rheological (MR) clutches offer a hardware-level solution for achieving both compliance and torque control through adjustable coupling between the input and output of the MR clutch. However, the presence of frequency-dependent magnetic hysteresis makes controlling the output torque challenging. In this paper, we present a comparative study of six widely used hysteresis models and propose a computationally efficient algebraic model to address the issue of hysteresis modeling and control of the output torque of rotary MR clutches. We compare the estimated torques with experimental measurements from a prototype MR clutch, to evaluate the computational complexity and accuracy of the model. Our proposed algebraic hysteresis model demonstrates superior accuracy and approximately two times less computational complexity than the Bouc–Wen model, and approximately twenty times less memory requirement than neural network-based models. We show that our proposed model has excellent potential for embedded indirect torque control schemes in systems with hysteresis, such as MR clutches and isolators.

show abstract

Section: Overview Of Hysteresis Modeling In Mr Clutchesmentioning

confidence: 99%

A Computationally Efficient Hysteresis Model for Magneto-Rheological Clutches and Its Comparison with Other Models

Yang

Kermani

2023

Actuators

View full text Add to dashboard Cite

show abstract

“…In accordance with the equation system’s structural characteristics in equation (15), the value of ε i · g ( θ ) will directly affect the accuracy of joint torsional stiffness identification, thus affecting the accuracy of positioning error identification and prediction. Since BP neural network has strong nonlinear mapping ability and good learning and adaptive ability, 23,24 this study uses BP neural network method to identify and predict the residual error based on the identification of joint torsional stiffness error model, in order to raise the robot’s flexible positioning precision. In general, if Sigmoid function 25 is adopted in the hidden layer and the number of hidden layer nodes is enough, a hidden layer can enable the neural network to approximate the required function with arbitrary accuracy.…”

Section: Flexible Deformation Error Model Of Robotmentioning

confidence: 99%

A calibration method for predicting and compensating the flexible positioning errors of a 7-DOF series compliant robot caused by end-effector’s gravity

Gao,

Xu,

Zhao

et al. 2023

Proceedings of the Institution of Mechanical Engineers, Part C:

View full text Add to dashboard Cite

Due to low stiffness of compliant series robot body, the influence of joint flexible deformation caused by the gravity load of the end-effector on the absolute positioning accuracy cannot be ignored. In order to improve the flexible positioning accuracy of the robot end-effector, a calibration method for predicting and compensating the flexible positioning error of the 7-DOF series compliant robot due to the gravity load of the end-effector is proposed. The method identifies and predicts the pose configuration with similar flexible positioning error through the semi-parametric error model based on joint stiffness identification and of BP neural network. The feedforward method is used to compensate the positioning error. In this study, a 7-DOF series compliant robot is selected as the research object. Based on the principle of error similarity, the measurement points and the corresponding pose configurations are selected as the training data for the parameter identification of the semi-parametric error model, and the flexible positioning errors on the continuous trajectory of the end-effector are compensated according to the predicted values of the model. When the mass of the end-effector is 5 kg, by this method, the maximum value, mean value and mean square error of flexible positioning error on multiple straight-line trajectories on the plane are increased from 8.4779, 5.6039, and 1.8496 mm to 0.9720, 0.3079, and 0.2303 mm, respectively. It means the positioning accuracy are increased by 88.53%, 94.50%, and 87.55%, respectively, by which could observe that the precision of this robot’s end-effector was significantly improved.

show abstract

“…Ma et al 6 proposed an improved dynamic linearization method to obtain an equivalent linear data model for learning systems. In recent years, learning-based control methods, such as reinforcement learning, [7][8][9][10] model predictive control, [11][12][13] and iterative learning control (ILC), 14,15 have rapidly developed. 16 ILC aims to continuously improve control performance by using periodic system response data through iterative methods.…”

Section: Introductionmentioning

confidence: 99%

An enhanced PI controller based on adaptive iterative learning control

Liang,

Yang,

et al. 2023

Intl J Robust & Nonlinear

View full text Add to dashboard Cite

SummaryThis paper presents a PI‐type adaptive iterative learning control (PI‐AILC) method for nonlinear processes, which targets enhancing system tracking capabilities by adapting setpoints of the PI controller. First, the proposed method employs compact form dynamic linearization technology to obtain a local linear representation of unknown nonlinear systems. Subsequently, the iterative learning controller gain is adaptively updated using the local linear expression to ensure the optimality of the setpoints. Finally, a pre‐learning mechanism for offline data is introduced to augment the efficiency of the iterative mechanism further. The proof of strict convergence for PI‐AILC is established. Experimental results substantiate the efficacy of PI‐AILC.

show abstract

Neural network based iterative learning control for magnetic shape memory alloy actuator with iteration-dependent uncertainties

Cited by 31 publications

References 41 publications

A Computationally Efficient Hysteresis Model for Magneto-Rheological Clutches and Its Comparison with Other Models

A Computationally Efficient Hysteresis Model for Magneto-Rheological Clutches and Its Comparison with Other Models

A calibration method for predicting and compensating the flexible positioning errors of a 7-DOF series compliant robot caused by end-effector’s gravity

An enhanced PI controller based on adaptive iterative learning control

Contact Info

Product

Resources

About