Research On CPG Control System Based On Flexible Joint In Six Legged Robot

Low biological rhythm movement is generated by the oscillation of the underlying central nervous system. Currently the motion control based on central pattern generator (CPG) has become one of the focus areas in bionic control CPG can produce movement of spontaneous, steady rhythm without any high-level external control signal and feedback, which doesn not require the model process on enviroment and itself. Thus, it can reduce the workload and save work time of work of the control system. But now the application of CPG is relatively limited, the application of the six legged robot is still remain in gait generation and swing phase control. The analysis shows that, CPG application in nature is the flexible organisms,but people used CPG in the rigid robots,therefore,the development of CPG bionic control is limited.In this paper, a control method is proposed which use CPG in the flexible joint robot, to make new attempt in the CPG application.First, analysis the actual joint motion characteristics of the six-legged insect, to provide the biological basis that CPG could control leg swing phase and support phase simultaneously. On this basis, focusing on one leg CPG network differential equations, study when coupling term is different, the different characters shown in the equations.Second, starting with the foot kinematics and mechanics of the robot, to keep stable in horizontal and vertical directions, the relationship and changes of the joint in motion is derivated.Third, hierarchical control system of hexapod robot is designed for the requirements of the control algorithm.Including integrated control and drive units, communication systems and sensor systems.Finally, The robot past the ground,uphil,downhill and unstructured road walking test to prove that the combination of CPG and flexible joints is effective, and successful control supporting phase using CPG, the application is expanded to the whole cycle in walking. It also shows the effectiveness and stability of the robot control system.