Study On CPG-Based Gait Planning For A Hexapod Bionic Robot

Hexapod robots possess broad application prospects for their various characteristics,such as high stability,various gait patterns,strong load capacity,etc.CPG coupled oscillator model can generate stable periodic oscillation signals spontaneously,and realize the coupling between multiple CPG units through constructing the topology network,which is suitable for gait planning of hexapod robots.First of all,the “Hex Ant” hexapod bionic robot was designed with reference to the body structure of the African red fire ant,and the freedom degree of the robot was analyzed from the mechanism.Hexapod robot kinematics model was established,which was simulated and verified by MATLAB.According to the rotation angle range of each joint,the working space of foot was simulated by MATLAB to determine the relation between the stroke size and working space of foot.Then,the basic concepts and the stability margin calculation formula of robot were determined.According to the number of supporting feet and the direction of walking,nine different gaits for the hexapod robot were planned out.By setting foot trajectory curve,each joint angle curve was obtained for different gaits and the stability margin of each gait could be calculated.The stability margin with the same gait and the different schemes were compared and analyzed to determine the optimal scheme of the gait.Next,the biological control mechanism of CPG was analyzed and applied to the robot's motion control.Method for CPG gait generation was determined.Comparing the characteristics of different types of oscillators,Hopf oscillator was chosen to establish the CPG model.Studies on convergence speed of limit cycles,amplitude,frequency,duty ratio for Hopf oscillator were carried out.The hierarchical structure of CPG network was set up,that the upper was used for coordinated control between legs and the lower was used for coordinated control between joints.The upper network has two kinds of connection modes: full symmetric network connection and the ring network connection.Under different connection modes,the output signals of CPG network with different gaits were obtained,and the advantages and disadvantages of two kinds of connection modes were compared.An appropriate network connection mode was selected according to the type of gait.Then,through the coupling between the legs and the coupling between joints in each leg,the coordinated and stable gait were realized.Finally,on account of triple supporting gait,quadruple supporting gait and quintuple supporting gait of “Hex Ant”,the robot experiment platform was set up to carry out nine gaits in straight,transverse and rotation directions respectively.The experimental results verified the effectiveness of the robot gait planning and CPG model.