Distributed Neural Dynamics Fault-Tolerance Scheme for Reliability-Synchronized Cooperative Motion Generation of Multirobot Manipulators

To improve the reliability and safety of multirobot manipulator systems (MRMSs) during the execution of cooperative tasks, a distributed neural dynamics fault-tolerance (DNDFT) approach is proposed and implemented. For this reason, combined with the kinematics of the robot manipulator, the performance loss malfunction model of joints is designed and given. Second, the synchronized motion generation constraints of the MRMSs are depicted as equality constraints with coupled variables based on the performance loss malfunction model of joints and the communication topology of leader–follower collaboration. In order to further enhance the reliability of MRMSs, the physical limits of the joints of each robot manipulator are considered and represented as inequality constraints. Then, the problem of reliability-synchronized cooperation of MRMSs is transformed into the quadratic programming form with the minimum velocity norm criterion, and the DNDFT approach is utilized to solve this problem. With the DNDFT approach, the multiple robot manipulators suffering from varying degrees of joint degradation can still satisfactorily perform the end-effector cooperative tracking task. Finally, computer simulations verify that the proposed DNDFT approach can effectively address the synchronized cooperative motion generation problem of MRMSs.