Article
Computer Science, Artificial Intelligence
Chenguang Yang, Dianye Huang, Wei He, Long Cheng
Summary: This article presents a control scheme for the robot manipulator's trajectory tracking task considering output error constraints and control input saturation. A bounded barrier Lyapunov function is proposed and adopted to handle the output error constraints. An auxiliary system is designed to suppress the input saturation effect, and a simplified RBFNN structure is adopted to approximate the lumped uncertainties. Simulation and experimental results demonstrate the effectiveness of the proposed control scheme.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Automation & Control Systems
Md Khurram Monir Rabby, Ali Karimoddini, Mubbashar Altaf Khan, Steven Jiang
Summary: This article proposes an adjustable autonomy framework for robot operation in human-robot collaboration scenarios. Through reinforcement learning and an integrated epsilon-greedy approach, the robot can autonomously adjust its actions based on rewards from a human operator. Experimental results confirm the effectiveness of the framework in various situations.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Robotics
Kourosh Darvish, Enrico Simetti, Fulvio Mastrogiovanni, Giuseppe Casalino
Summary: FlexHRC+ is a hierarchical human-robot cooperation architecture designed to provide collaborative robots with extended autonomy, including three levels for perception, representation, and action. The key features include a decision-making process for collaborative robots' operations and a representation level with behavior formally specified using first-order logic.
IEEE TRANSACTIONS ON ROBOTICS
(2021)
Article
Automation & Control Systems
Yongqing Fan, Zhan Zhu, Zhen Li, Chenguang Yang
Summary: This paper studies an adaptive radial basis function neural network (RBFNN) control scheme for desired tracking of multiple robot manipulators carrying a common object in joint-space. The paper introduces a non-zero time-varying parameter into the RBFNN and presents a novel universal approximation of RBFNN with this parameter. A switching control scheme is designed based on this approximation property to handle uncertainties and improve control effectiveness. The effectiveness of the proposed scheme is demonstrated through simulation results.
EUROPEAN JOURNAL OF CONTROL
(2023)
Article
Automation & Control Systems
Tianhao Qie, Xinan Zhang, Chaoqun Xiang, Yang Yu, Herbert Ho Ching Iu, Tyrone Fernando
Summary: This article proposes an online integral reinforcement learning (IRL) based data-driven control algorithm for interleaved dc/dc boost converter. The algorithm uses a three-layer neural network (NN) and is independent of the system model. The controller gains are autonomously adjusted online through the value function based NN weights updating mechanism, simplifying the controller gain tuning process. Compared to conventional model-dependent control approaches, it provides superior control performance. Moreover, the proposed method significantly reduces the computational burden of classical IRL algorithm by removing the disturbance updating process. Experimental results verify the efficacy of the proposed algorithm.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2023)
Article
Automation & Control Systems
Yingying Li, Yujie Tang, Runyu Zhang, Na Li
Summary: This article proposes a zero-order distributed policy optimization algorithm (ZODPO) for a distributed reinforcement learning problem in decentralized linear quadratic (LQ) control. The algorithm leverages the ideas of policy gradient, zero-order optimization, and consensus algorithms to learn linear local controllers in a distributed manner. Experimental results demonstrate the scalability and control stability of the algorithm.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2022)
Article
Engineering, Electrical & Electronic
Shuyang Liu, Siyuan Tong, Yuanchun Li
Summary: This article studies an adaptive boundary control strategy for a rigid-flexible robot with concurrence failures and multiple constraints. By designing a boundary controller and using fault-tolerant control to compensate for the effects of failures, position tracking and vibration suppression are achieved.
IEEE SENSORS JOURNAL
(2023)
Article
Automation & Control Systems
Dianye Huang, Chenguang Yang, Yongping Pan, Long Cheng
Summary: This article introduces a control scheme for robot manipulators that takes into account output error constraints, unknown dynamics, and bounded disturbances. By proposing a modified virtual input variable and implementing composite learning laws for enhancing neural networks, the controller's robustness is improved. Experimental results demonstrate the superiority of the proposed controller in terms of parameter estimation and tracking capabilities.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Automation & Control Systems
Ziyu Chen, Yang Liu, Wei He, Hong Qiao, Haibo Ji
Summary: This article presents an adaptive neural network control scheme for an uncertain wheeled mobile robot with velocity constraints and nonholonomic constraints. The scheme utilizes adaptive neural networks to approximate unknown robotic dynamics and employs barrier Lyapunov function to ensure velocity constraints, with simulations and practical experiments demonstrating its effectiveness.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2021)
Article
Automation & Control Systems
Anna Scampicchio, Aleksandr Aravkin, Gianluigi Pillonetto
Summary: The study addresses the finite-horizon and discrete-time LQR problem with stability constraints and uncertain system dynamics, modeling the robustness of the solution using the scenario approach and applying the methods to the Leslie growth model for population control.
Article
Engineering, Aerospace
Han Wu, Qinglei Hu, Yongxia Shi, Jianying Zheng, Kaipeng Sun, Jiawen Wang
Summary: This paper addresses the optimal impedance control problem for large-scale space manipulators with unknown contact dynamics and partial measurements. A novel model-free value iteration integral reinforcement learning algorithm is proposed to approximate optimal impedance parameters. The proposed algorithm eliminates the need for prior contact dynamics knowledge and full-state measurements while reducing the dependence on specific initial stabilization control.
AEROSPACE SCIENCE AND TECHNOLOGY
(2023)
Article
Automation & Control Systems
Karla Rincon, Isaac Chairez, Wen Yu
Summary: This study introduces a new type of trajectory tracking robust controllers for a class of rehabilitation robotic system considering articulations restrictions. Numerical evaluations confirm the effectiveness of the proposed controller in tracking reference trajectories and satisfying articulation restrictions.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2022)
Article
Automation & Control Systems
Shan Xue, Biao Luo, Derong Liu, Ying Gao
Summary: An event-triggered adaptive dynamic programming (ADP) algorithm is proposed for tracking control of partially unknown constrained uncertain systems. The algorithm constructs an augmented system, employs integral reinforcement learning, and utilizes event triggering to relax initial control requirements and ensure bounded tracking and weight estimation errors. Simulation results demonstrate the effectiveness of the approach.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Automation & Control Systems
Yunxia Song, Wim Michiels, Bin Zhou, Guang-Ren Duan
Summary: This article addresses the strong stability problem of linear continuous-time delay-difference equations with multiple time delays. It derives a family of linear matrix inequalities indexed by a positive integer k to assess strong stability, and provides a time-domain interpretation in terms of a quadratic integral Lyapunov functional to reveal relations with existing results. The LMI condition can be easily reformulated to establish a sufficient condition for robust strong stability and a necessary and sufficient condition is also given in the form of a structured singular value characterization.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2021)
Article
Automation & Control Systems
Jin-Gang Zhao, Fang-Fang Chen
Summary: This article explores the optimal tracking control problem of a class of multi-input nonlinear systems with unknown dynamics using reinforcement learning and nonzero-sum game theory. Various algorithms are proposed and analyzed to approximate the Nash equilibrium solution, as well as to eliminate the need for prior knowledge of system dynamics through off-policy integral reinforcement learning implemented by neural networks. The effectiveness of the proposed methods is demonstrated through numerical simulations.
OPTIMAL CONTROL APPLICATIONS & METHODS
(2022)