Virtual dynamic modelization, control and identification of bipedal robots

This paper develops a new dynamic virtual model of bipedal robot. The key in this generic equational representation is the use of recent developments in manipulators and inverted pendulum kind models robots. It gives a mathematical based control strategy. This control strategy permits the biped motion in a deterministic environment. It concerns the on-line pursuit with two phases based programs: the minimum time ballistic phase and the adaptive feedback approach phase. The MTBP-AFAP programs allow the realization of the whole task of walking from a point A to a point B adaptively and time optimally