ADRC Based Multi-task Priority Tracking Control for Collaborative Robots

Authors: Fan, K., Liu, Y., Zhang, K., Bian, G. and Yu, H.

Journal: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Volume: 13606 LNAI

Pages: 439-450

eISSN: 1611-3349

ISSN: 0302-9743

DOI: 10.1007/978-3-031-20503-3_35


When collaborative robots perform multiple tracking tasks at the same time, the dynamics of each task will interact with each other. In addition, the uncertainty of robot model and external perturbation also limit the performance of the system. In order to tackle above problems, a task priority control framework combining ADRC and null space projection is proposed. First, by using the null-space projection, the new state variables which can realize the decoupling of task space inertia are obtained and the task space dynamics is inertially decoupled correspondingly. Then, by establishing the relationship between the new task state variables and the original task state variables, the task space dynamics represented by the original task state variables is got. Thirdly, based on the task space dynamics of each task, a ADRC controller is constructed to improve the tracking performance and disturbance attenuation property simultaneously. Simulation experiments on the UR5 robot verify the effectiveness of the proposed controller.

Source: Scopus