Inverse Kinematics Solution Based on Redundancy Modeling and Desired Behaviors Optimization for Dual Mobile Manipulators

Dual mobile manipulators are increasingly used in cooperative operation tasks, and their appropriate joint configurations are the premise to ensure the success of the tasks. However, for a given task path, due to the existence of redundant degrees of freedom, there are countless possible joint configurations, which makes it difficult to solve the inverse kinematics. In addition, most of the existing redundant inverse kinematics solutions are only for redundant manipulators or single mobile manipulator, with less consideration given to the impact of the constraints between the dual mobile manipulators on the solution performance. To this end, a new inverse kinematics solution method for the dual mobile manipulators is proposed in the paper. First, six parameters are introduced to describe the kinematic redundancy, so as to transform the inverse kinematics of the dual mobile manipulators into non-redundant inverse kinematics. Then, in order to make full use of the redundancy, the weighted product form of the normalized ideal redundant behaviors including manipulability, joint motion continuity, collision avoidance, obstacle avoidance and joint motion synchronization is considered as an objective function. Finally, the redundancy parameters are optimized by the Nelder-Mead simplex method to obtain the desired joint configurations. Based on the platform composed of SUMMIT-XL STEEL mobile robot and UR5 manipulator, a single target pose solution is simulated. Moreover, using the platform composed of SUMMIT-XL STEEL mobile robot and AUBO-i16 manipulator, the performance of path solution and obstacle avoidance is evaluated experimentally. The simulation and experimental results show that under the constraints of the desired redundant behaviors, satisfactory joint configurations can be acquired in the reachable space of the joints. Besides, the joint configurations of the dual mobile manipulators are solved in about 1 ms. It follows that the proposed method can select the optimized one from multiple possible joint configurations by the redundancy parameters optimization.