Adaptive Sliding-Mode Disturbance Observer-Based Nonlinear Control for Unmanned Dual-Arm Aerial Manipulator Subject to State Constraints

Abstract

The unmanned dual-arm aerial manipulator system is composed of a multirotor unmanned aerial vehicle (UAV) and two manipulators. Compared to a single manipulator, dual-arm always provides greater flexibility and versatility in both goods delivery and complex task execution. However, the practical application of the system is limited due to nonlinearities and complex dynamic coupling behavior between the multirotor and the manipulator, as well as the one between the inner and outer loop of the multirotor. In this paper, a holistic model of the dual-arm aerial manipulator system is first derived with complete model information. Subsequently, an adaptive sliding-mode disturbance observer (ASMDO) is proposed to handle external disturbances and unmeasurable disturbances caught by unmeasurable angular velocity and acceleration of the manipulators. Moreover, for safety concerns and transient performance requirements, the state constraints should be guaranteed. To this end, an auxiliary term composed of constrained variable signals is introduced. Then, the performance of the designed method is proven by rigorous analysis. Finally, the proposed method is validated through two sets of simulation tests.