Redundancy-Driven Multi-Task Adaptive Backstepping Tracking Control for Aerial Manipulators

This paper presents a novel multi-tasking control scheme for an aerial manipulator consisting of an unmanned aerial vehicle (UAV) and a robotic arm as a high degree-of-freedom (DOF) system. The decoupled dynamic model is investigated under uncertainties to precisely control the motion of the UAV and the robotic arm. To reduce unnecessary motion of the end-effector, the null-space behavioral (NSB) strategy is utilized to perform subtasks. This feature provides a smoother trajectory for the transported object. The convergence is theoretically analyzed by utilizing Lyapunov stability. The proposed control scheme is validated with numerical simulations and experiments in several scenarios. To verify the efficacy of the proposed method, two types of subtasks, joint angle limitation (JAL) and obstacle avoidance (OA), are presented to demonstrate the effectiveness of multi-tasking. Finally, experimental results for collision avoidance are provided to verify that the system can be implemented in practice. With the device's inherent noise, the root-mean-square error remains at approximately 5 cm for the UAV frame ZD850.