A null space coordination algorithm for mobile manipulator end-point steady control problem

This paper presents a whole-body control algorithm designed to enhance both safety and accuracy in end effector control. Initially, a saturation null space algorithm is introduced, enabling the base and end effector to execute tasks with varying priorities while adhering to inequality safety constraints. To tackle the issue of low base positioning accuracy impacting the end effector’s precision in real-world applications, a localization strategy integrating probabilistic localization with point cloud matching is implemented. Subsequently, a series of simulations and real-world experiments were conducted to validate the system’s capability to precisely adhere to various equality and inequality constraints, including trajectory and speed constraints. These experiments demonstrated that the system sustained an end effector control precision of 0.006m in real-world environments, unaffected by indoor localization errors. Furthermore, when combined with a 2D laser scanner-based indoor localization algorithm, the system’s control precision was maintained at 0.02m, achieving a reduction in error of over 50% compared to using AMCL for localization directly.