Adaptive robust control of multi-DOF hydraulic manipulator with precise online load estimation

Hydraulic manipulator shows vast application potential in heavy load working conditions. However, achieving high control precision in these devices is notably more challenging compared to electric manipulators, due to the complexities introduced by uncertainties and high-order nonlinear dynamics. The presence of heavy unknown payload further reduces control accuracy. In this paper, a new load estimation method based on direct/indirect adaptive robust controller (DIARC) is proposed to facilitate online payload estimation and compensate the impact of the unknown payload. To compensate for the high-order dynamics, backstepping strategy is utilized. A modified recursive least squares adaptive law is also developed to realize, precise, real-time payload estimation under dynamic conditions. By incorporating this accurate load mass estimation into the control strategy, an improvement in overall control performance can be achieved. The effectiveness of this enhanced controller with load mass estimation is initially verified through simulations conducted in MATLAB. The close-loop control performance is further analyzed and validated on a four-degree-of-freedom hydraulic manipulator. The experimental results indicate that, under dynamic scenarios, our proposed control method succeeds in achieving precise online load mass estimation, achieving an average estimation error of 2.8% for a 7.5 kg payload. Furthermore, enhanced control accuracy is achieved compared to traditional controllers, with the maximum tracking error being only 0.69° during the simulated operation scenario. This research provides a viable solution for precision control and load estimation in hydraulic manipulators, eliminating the need for costly force/torque sensors.