An improve nonlinear robust control approach for robotic manipulators with PSO-based global optimization strategy

Abstract

During the trajectory tracking of robotic manipulators, many factors including dead zones, saturation, and uncertain dynamics, greatly increase the modeling and control difficulty. Aiming for this issue, a nonlinear active disturbance rejection control (NADRC)-based control strategy is proposed for robotic manipulators. In this controller, an extended state observer is introduced on basis of the dynamic model, to observe the extend state of model uncertainties and external disturbances. Then, in combination with the nonlinear feedback control structure, the robust trajectory tracking of robotic manipulators is achieved. Furthermore, to optimize the key parameters of the controller, an improved particle swarm optimization algorithm (IPSO) is designed using chaos theory, which improves the tracking accuracy of the proposed NDRC strategy effectively. Finally, using comparative studies, the effectiveness of the proposed control strategy is demonstrated by comparing with several commonly used controllers.

Keywords: Active disturbance rejection controller; Nonlinear control; Nonlinear dynamics; Particle swarm optimization; Robotic manipulator.

Fig. 1

Structure of 6-degree of freedom robotic arm and its simplified schematic diagram.

Fig. 2

Optimization framework of ADRC based on the improved PSO algorithm.

Fig. 3

Structure of the active disturbance rejection controller.

Fig. 4

Structure of the active disturbance rejection controller.

Fig. 5

Comparison results for Fitness values of the test function (10 dimensions).

Fig. 6

Comparison results for Fitness values of the test function (20 dimensions).

Fig. 7

Variation of fitness values with the increasing of iterations.

Fig. 8

tracking result on step reference signals.

Fig. 9

Tracking result on reference signals.

Fig. 10

Tracking errors on reference signals.

Fig. 11

Control inputs of different controllers.