Search results

This paper aims to study a parameter tuning method for the active disturbance rejection control (ADRC) to improve the anti-interference ability and position tracking of the performance of the servo system, and to ensure the stability and accuracy of practical applications.

This study proposes a parameter self-tuning method for ADRC based on an improved glowworm swarm optimization algorithm. The algorithm is improved by using sine and cosine local optimization operators and an adaptive mutation strategy. The improved algorithm is then used for parameter tuning of the ADRC to improve the anti-interference ability of the control system and ensure the accuracy of the controller parameters.

The authors designed an optimization model based on MATLAB, selected examples of simulation and experimental research and compared it with the standard glowworm swarm optimization algorithm, particle swarm algorithm and artificial bee colony algorithm. The results show that the response time of using the improved glowworm swarm optimization algorithm to optimize the auto-disturbance rejection control is short; there is no overshoot; the tracking process is relatively stable; the anti-interference ability is strong; and the optimization effect is better.

The innovation of this study is to improve the glowworm swarm optimization algorithm, propose a sine and cosine, local optimization operator, expand the firefly search space and introduce a new adaptive mutation strategy to adaptively adjust the mutation probability based on the fitness value, improve the global search ability of the algorithm and use the improved algorithm to adjust the parameters of the active disturbance rejection controller.

To control one of the joints during the actual movement of the hydraulically driven quadruped robot, all the other joints in the leg need to be locked. Once the joints are unlocked, there is a coupling effect among the joints. Therefore, during the normal exercise of the robot, the movement of each joint is affected by the coupling of other joints. This brings great difficulties to the coordinated motion control of the multi-joints of the robot. Therefore, it is necessary to reduce the influence of the coupling of the hydraulically driven quadruped robot.

To solve the coupling problem with the joints of the hydraulic quadruped robot, based on the principle of mechanism dynamics and hydraulic control, the dynamic mathematical model of the single leg mechanism of the hydraulic quadruped robot is established. On this basis, the coupling dynamics model of the two joints of the thigh and the calf is derived. On the basis of the multivariable decoupling theory, a neural network (NN) model reference decoupling controller is designed.

The simulation and prototype experiment are carried out between the thigh joint and the calf joint of the hydraulic quadruped robot, and the results show that the proposed NN model reference decoupling control method is effective, and this method can reduce the cross-coupling between the thigh and the calf and improve the dynamic characteristics of the single joint of the leg.

The proposed method provides technical support for the mechanical–hydraulic cross-coupling among the joints of the hydraulic quadruped robot, achieving coordinated movement of multiple joints of the robot and promoting the performance and automation level of the hydraulic quadruped robot.

On the basis of the theory of multivariable decoupling, a new decoupling control method is proposed, in which the mechanical–hydraulic coupling is taken as the coupling behavior of the hydraulic foot robot. The method reduces the influence of coupling of system, improves the control precision, realizes the coordinated movement among multiple joints and promotes the popularization and use of the hydraulically driven quadruped robot.