Generation of bipedal gait suitable for use in unstructured environments

When walking in an unstructured environment, human is realizing various trajectories, often arising due to an instantaneous situation that could not be predicted in advance. This imposes the requirement for on-line trajectory planning and constant changing of gait parameters (turning, stopping, acceleration and deceleration, switching from the walk on a flat to the walk on an inclined surface or staircases, etc.). In view of the prospect that humanoid robots of the future will share with humans living and working space it will be normally expected from them to possess similar walking capabilities. This work aims to demonstrate the possibility of on-line generation of complex motion as a combination of simple motions at particular joints, called primitives. Each primitive has its parameters and constraints that are determined on the basis of similar man's motions. The set of all primitives represents the base from which primitives are selected and combined for the purpose of performing the corresponding complex movement. The proof that a correct selection of primitives is made and that the movement is the appropriate one is obtained on the basis of the maintainance of dynamic balance, which is realized by monitoring the ZMP position, as well as based on the pattern of the very movement.