Active/passive walking strategy for a biped robot using CPG with sensory interaction

For the locomotion control of biped robot in active/passive hybrid walking, a Central Pattern Generator based motion control strategy is proposed. The strategy uses CPG as equations with respect to the state of supporting leg, and generates proper limb movements to perform the walking gait for biped robot. The orientation angle of the limbs is used as the inner output of rhyme oscillator. Sensors input including ground reaction force and state of trunk modulate the CPG output among the biped walking. Active and passive walking phase is analyzed. And configuration of the motion generator is discussed. Then the biped walking strategy by using CPG is established. A simulation model containing collisions is established. Biped walking simulation is conducted, and the motion control system could generate a stable biped walking. Simulation results show the effectiveness of the proposed strategy.