Motion/force control scheme for electrically driven cooperative multiple mobile manipulators

Adaptive Reinforcement Learning-Enhanced Motion/Force Control Strategy for Multirobot Systems

“…Among different schemes, robotic systems can mainly be separated into two categories: (1) mobile robotics [3,4] and (2) robot manipulators [5]. However, because of the growing requirements for high productivity, the networked robotic systems that are known as the collaboration of many robots can fulfill different tasks by mobile manipulators [6], tractor trailers [3], bilateral teleoperators [5], and cooperating mobile manipulators (CMMs) [7]. It is worth emphasizing that, unlike nonlinear multiagent systems [8,9] studying the relation between agents, CMMs consider the interaction with the rigid object.…”

“…It is worth emphasizing that, unlike nonlinear multiagent systems [8,9] studying the relation between agents, CMMs consider the interaction with the rigid object. In the literature of CMMs' control objectives, they can be classified into two categories, i.e., multiple mobile robot manipulators in cooperation carrying a common object with unknown parameters and disturbances [7,[10][11][12][13][14][15][16][17][18]; one of them tightly holds the object by the end effector, and the remaining mobile manipulators' end effector follows a trajectory on the surface of the object [19,20]. Most of the existing control literature of networked robotics mainly focuses on the implementation of nonlinear controllers.…”

“…As a result, the robust adaptive controller was realized by estimating the derivative of the Lyapunov function and combining the constraint force control to obtain the motion/force control scheme for multimanipulator systems [12]. Due to dynamic uncertainties and external disturbance, fuzzy and neural network (NN) method are popularly considered to estimate dynamic uncertainties for guaranteeing system stability [7,10,20]. However, the work in [23] utilized an estimation in the absence of fuzzy and NNs for designing a decentralized controller for cooperating manipulators.…”

“…However, the work in [23] utilized an estimation in the absence of fuzzy and NNs for designing a decentralized controller for cooperating manipulators. In contrast to this technique, authors in [7,12,25] proposed the centralized control law based on the consideration of dynamics of m manipulators. e scheme in [7] extends the neural network-based centralized coordination control to achieve the tracking of each mobile manipulator by the consideration of additional actuators using servo motors.…”

“…In contrast to this technique, authors in [7,12,25] proposed the centralized control law based on the consideration of dynamics of m manipulators. e scheme in [7] extends the neural network-based centralized coordination control to achieve the tracking of each mobile manipulator by the consideration of additional actuators using servo motors. However, these nonlinear control solutions do not take into account the challenges, such as input-state saturation, and control objectives are different from the traditional tracking control problem.…”