Abstract
This paper proposes an optimal path planning for unmanned ground vehicle considering kinematics and dynamics of a vehicle model. We set phase space including posture and velocity of the vehicle and use RASMO (Rapid Semi-Optimal Motion-Planning)kim2010 to generate search trees in four dimensional phase space. In this paper, we calculate transition cost based on not only running time but also simple cost map which represents road surface condition and we choose the path with the minimal cost by the search trees. It was verified that the proposed method can generate the path including complex movements like a cuff to avoid bad condition area.
