Motion control of hydraulic quadruped robot based on central pattern generator

Using the animals neural control mechanism in robot motion control is a rapidly growing research direction. Biocybernetics method of the robot is established on the basis of simulating, simplifying and improving the control area of animal rhythmic motion (CPG-central pattern generator, high-level control central, biological reflection, etc.). Rhythm motion can improve the performance of the robot in the actual environment. From an engineering perspective, CPG neural network can be viewed as a distribution system which is comprised of a group of mutually coupled nonlinear oscillators. Changing the coupling relationship of oscillators can generate time sequence signals with different phase relationships and achieve different motion modes. This paper researches the use of rhythmic motion to achieve the motion control of quadruped robot. The method of Matsuoka is used to establish central pattern generator (CPG) model and setting parameters. A wholly symmetrical quadruped robot CPG network with the concept of rhythmic gait can achieve the mutual conversion between a series of typical motion gaits.