Quadruped robot single-leg trajectory tracking control method and system

Abstract

The invention provides a quadruped robot single-leg trajectory tracking control method and system. The quadruped robot single-leg trajectory tracking control method comprises the steps of obtaining joint parameters of a quadruped robot; establishing a single-leg dynamic model according to the joint parameters; correcting the single-leg dynamical model based on disturbance and a neural network, constructing a single-leg dynamical model under the condition of considering disturbance, and assuming a reference trajectory; obtaining an optimal parameter of the reference trajectory by using a value function approximation method based on neural network learning, and obtaining an optimal control rate; and achieving single-leg trajectory tracking control of the quadruped robot based on the optimal control rate. The method solves the problem of how to compensate the influence of nonlinearity and errors, improves the performance of a nonlinear control system of the quadruped robot, guarantees that the trajectory tracking error of the quadruped robot is gradually converged to zero, and achieves the high-precision trajectory tracking control of the quadruped robot.

Description

technical field [0001] The disclosure belongs to the technical field of robot control, and in particular relates to a single-leg trajectory tracking control method and system for a quadruped robot. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art. [0003] Since the end of the last century, the research on robotics has been a research hotspot in scientific research institutions and technology companies at home and abroad. Footed bionic robots are one of the most challenging topics in robotics. Among them, quadruped robots are the most developed in the field of bionic robots. One of the fastest robot types, quadruped robots have the characteristics of high dynamic performance, strong adaptability to complex terrain, and strong anti-interference ability. Compared with wheeled robots, they are more flexible and have stronger passability and Adaptabil...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the above problems, this disclosure aims at the single leg of the quadruped robot, and uses the excellent nonlinear approximation ability of the neural network to propose a quadruped robot single leg based on RBF (Radial Basis Function) neural network (NN, Neural Networks) compensation control. The leg trajectory tracking control method is used to compensate the influence of nonlinearity and error, improve the performance of the nonlinear control system of the quadruped robot, ensure that the trajectory tracking error of the quadruped robot gradually converges to zero, and realize the high-precision trajectory tracking control of the quadruped robot ; It is feasible to perform trajectory tracking control with neural network compensation for the single leg of a quadruped robot. Based on neural network control technology, many complex system control problems can be solved. These complex system control problems include robots and manipulator systems.

Images