Time optimal path planning while considering physical and geometrical limits during obstacle avoidance

In this paper, an approach for a wheeled mobile robot to deal with the geometrical limits of a given path while also satisfying its physical constraints is presented. Presence of obstacles along a path is inevitable in practical mobile robot navigation. To avoid incoming collision, the robot is redirected to a new path where high-curvature turning points are addressed to be imminent. A known static obstacle is placed within a pre-determined Bézier curve-based path for the robot to exhibit obstacle avoidance accordingly. A trajectory generated through the robot’s acceleration limits according to the path’s curvature, expressed lengthy travel time, non-periodic time sampling, and does not consider velocity limits configured on the robot. To deal with prevailing issues, a convolution-based approach was implemented but it demonstrated inability to track the path of high curvature in uniform sampling time. Therefore, a variation of linear interpolation was proposed to generate an actual trajectory in tracking the redirected path. The proposed time optimal path planning is able to consider the robot’s physical limits, cope with the path’s geometrical constrictions and conserve uniform time sampling to be conducted in real-time control.