Trajectory prediction and path planning of intelligent autonomous biped robots - learning and decision making through perception and spatial reasoning

Abstract

This paper is inspired by FP6 IST RobotCub European research project. The paper is addressed to the synthesis of an intelligent autonomous locomotion (artificial gait) of biped robots in unknown and unstructured dynamic environments through perception, learning, environment understanding and spatial reasoning. Focusing the research activities to the embodied cognition, this paper contributes to the extension of the intelligent robot behavior through dynamic environment understanding, simultaneous localization and mapping, trajectory prediction and path planning, obstacle avoidance, collision avoidance and scenario-based behavior. The paper is of a conceptual character. It includes some simulation results obtained by application of fuzzy algorithms and artificial neural networks in learning and decision making.