Biped robots have better mobility than\nconventional wheeled robots. The bio-inspired method\nbased on a central pattern generator (CPG) can be\nused to control biped robot walking in a manner like\nhuman beings. However, to achieve stable locomotion,\nit is difficult to modulate the parameters for the neural\nnetworks to coordinate every degree of freedom of the\nwalking robot. The zero moment point (ZMP) method\nis very popular for the stability control of biped robot\nwalking. However, the reference trajectories have low\nenergy efficiency, lack naturalness and need significant\noffline calculation. This paper presents a new method\nfor biped real-time walking generation using a hybrid\nCPG-ZMP control algorithm. The method can realize a\nstable walking pattern by combining the ZMP criterion\nwith rhythmic motion control. The CPG component is\ndesigned to generate the desired motion for each robot\njoint, which is modulated by phase resetting according\nto foot contact information. By introducing the ZMP\nlocation, the activity of the CPG output signal is adjusted\nto coordinate the limbs� motion and allow the robot to\nmaintain balance during the process of locomotion. The\nnumerical simulation results show that, compared with\nthe CPG method, the new hybrid CPG-ZMP algorithm\ncan enhance the robustness of the CPG parameters\nand improve the stability of the robot. In addition, the proposed algorithm is more energy efficient than the\nZMP method. The results also demonstrate that the\ncontrol system can generate an adaptive walking pattern\nthrough interactions between the robot, the CPG and the\nenvironment.
Loading....