A bio-inspired postural control for a quadruped robot : an attractor-based dynamic

Postural stability is a requirement for autonomous adaptive legged locomotion. Neurobiological research lead to the idea that there are independent central systems for posture and locomotion, which interact when required. In this work we propose a posture control system focused in the standing posture context. We integrate the proposed posture system with a CPG design based on coupled nonlinear oscillators. The proposed system generates movements for posture correction which are modulated according to sensory information. We integrate several different responses that individually contribute to the posture equilibrium. This coordination, competition and redundancy among the responses is a key element for adaptive, flexible and fault tolerant motor system. The control system is validated through a few experiments, where the robot is subjected to different posture situations ranging from roll and pitch variations to loss of feet support