A Coordinated Planning and Control Framework for Mobile Dual-arm Robots with Manipulability Optimization and Force Tracking

Abstract

Mobile dual-arm robots, characterized by their high degree of freedom, often encounter challenges such as kinematic singularities and motion/force control during mobile manipulation tasks. These challenges can constrain their manipulation performance and control precision. In this paper, we first develop a manipulability optimization strategy for mobile dual-arm robots and equivalently transform it into a convex quadratic programming framework. This transformation enables optimal trajectory generation within the constraints of joint physical limits and trajectory tracking requirements, thereby optimizing manipulability while avoiding singularity issues. Subsequently, we propose a hybrid control strategy based on a master-slave control framework. This strategy employs a selection matrix to coordinate motion and contact force tracking, effectively adapting to the varying demands of motion/force control during different phases of manipulation. Furthermore, we introduce a variable stiffness impedance control strategy to enhance the precision of internal force tracking. Finally, we implement a mobile manipulation task on a simulation platform. This coordinated planning and control framework is the first to combine manipulability optimization and contact force tracking on a mobile dual-arm robot. The simulation results demonstrate that our approach significantly enhances manipulation performance and improves force tracking precision compared to existing strategies. Specifically, the average manipulability increases by approximately 40%, and the steady-state error in internal force tracking is reduced from 1.84N to 1.08N. This validates the effectiveness and advantages of the proposed planning and control scheme.