Critic Only Policy Iteration-based Zero-sum Neuro-optimal Control of Modular and Reconfigurable Robots with uncertain disturbance via Adaptive Dynamic Programming

An, Tianjiao; Chen, Jingchen; Zhu, Xinye; Li, Yuanchun; Liu, Keping; Dong, Bo

doi:10.1109/icaci49185.2020.9177637

Cited by 4 publications

(4 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two different control methods are used in the simulations that contain the existing learning-based control scheme, such as An et al (2020), An et al (2021), Dong et al (2019), and the proposed position–force–based zero-sum approximate optimal control method. In the figures, (a) represents the joint 1 as well as (b) denotes the joint 2.…”

Section: Simulationmentioning

confidence: 99%

“…In order to improve the robustness, a neuro-optimal method is proposed in Kong et al (2021). Zhu et al (2020) developed position-force optimal control of manipulator under uncertain disturbance.…”

Section: Introductionmentioning

confidence: 99%

“…Optimal control method has been proposed nearly 70 years ago (Bellman and Dreyfus, 2013; Pontryagin, 1959), and plenty of literatures are recorded (An et al, 2021; An et al, 2021; Wei et al, 2021). Adaptive dynamic programming (ADP) theory is a key direction to undertake approximate optimal control system issues of discrete-time (Hu et al, 2021; Luo et al, 2020; Zhu et al, 2020a), continuous-time (An et al, 2021; Shan et al, 2020; Wei et al, 2020), data driven (Gao et al, 2018; Li et al, 2020; Su et al, 2020), and furthermore robot systems with input/output constraints (An et al, 2020; Lu et al, 2020; Ren et al, 2019), external disturbance (An et al, 2019; Song and Lewis, 2020; Xia et al, 2020), actuator failures (Dai et al, 2018; Jiao et al, 2018; Ma et al, 2020), and so on. Kong et al (2021b) developed n -link manipulator approximate optimal law via saturation nonlinearity.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Decentralized position–force zero-sum approximate optimal control for reconfigurable robots with unmodeled dynamic

Zhu

Wang

2022

Transactions of the Institute of Measurement and Control

Self Cite

View full text Add to dashboard Cite

In this paper, the position–force–based approximate optimal control method is developed for reconfigurable robots using zero-sum game strategy. By utilizing the Newton–Euler iteration technique, the robotic system’s dynamic model is formulated and the state space equation is derived. According to adaptive dynamic programming (ADP) and neural network algorithm, the trajectory tracking control problem is transformed into a zero-sum game-based optimal control issue. The optimal control policy and worst disturbance policy are obtained by Hamilton–Jacobi–Issacs (HJI) function, respectively. Unlike the conventional learning–based robotic control method, the proposed zero-sum game-based method no need extra sub-controller that can reduce the computational burden. The reconfigurable robot system’s tracking error is uniformly ultimately bounded by the Lyapunov theorem. Finally, simulation experiments demonstrate the advantages of the proposed method.

show abstract

Section: Simulationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Decentralized position–force zero-sum approximate optimal control for reconfigurable robots with unmodeled dynamic

Zhu

Wang

2022

Transactions of the Institute of Measurement and Control

Self Cite

View full text Add to dashboard Cite

show abstract

“…In practical applications, uncertain interference such as continuous environmental contact will result in the deterioration of robot performance. Zero-sum game 23,24 is proposed to solve the approximate optimal control problem. An et al 25 presented a zero-sum game method via policy iteration (PI).…”

Section: Introductionmentioning

confidence: 99%

Event‐triggered adaptive fuzzy optimal control of modular robot manipulators using zero‐sum differential game through value iteration

Dong

Feng

Cui

et al. 2023

Adaptive Control & Signal

View full text Add to dashboard Cite

SummaryIn this article, a novel event‐triggered zero‐sum differential game through value iteration (VI)‐based adaptive fuzzy control is proposed for modular robot manipulators (MRMs) with uncertain disturbance. The subsystem dynamic of MRMs is established by joint torque feedback (JTF) technology. An identifier based on fuzzy logic model (FLM) is established to identify unknown system dynamics. Then, the trajectory tracking control issue of MRMs with uncertain disturbance is transformed into a two‐player zero‐sum differential game. On the basis of event‐triggered mechanism and VI algorithm, the event‐triggered Hamilton–Jacobi–Issacs (HJI) equation is approximately solved through establishment of actor‐critic structure. Different from the traditional adaptive dynamic programming (ADP) algorithm, this scheme is based on adaptive fuzzy optimal control through an actor‐critic structure, which refers to adaptation of both optimal cost function and optimal control policy. Moreover, tracking errors of closed‐loop MRM system are demonstrated to be uniformly ultimately bounded (UUB) under Lyapunov stability theorem. At last, experimental examples and comparisons show the reliability and effectiveness of this method.

show abstract

Optimal H∞ tracking control of nonlinear systems with zero-equilibrium-free via novel adaptive critic designs

Peng

Zou

et al. 2023

Neural Networks

View full text Add to dashboard Cite

Critic Only Policy Iteration-based Zero-sum Neuro-optimal Control of Modular and Reconfigurable Robots with uncertain disturbance via Adaptive Dynamic Programming

Cited by 4 publications

References 12 publications

Decentralized position–force zero-sum approximate optimal control for reconfigurable robots with unmodeled dynamic

Decentralized position–force zero-sum approximate optimal control for reconfigurable robots with unmodeled dynamic

Event‐triggered adaptive fuzzy optimal control of modular robot manipulators using zero‐sum differential game through value iteration

Optimal H∞ tracking control of nonlinear systems with zero-equilibrium-free via novel adaptive critic designs

Contact Info

Product

Resources

About