Path planning for line marking robots using 2D Dubins' path

Abstract

This paper proposes a path planning algorithm based on 2D Dubins’ path for the construction of a curvature continuous trajectory for the autonomous guidance of line marking robots in football stadiums. The algorithm starts with four corner points representing the playable football field and generates a set of waypoints representing various parts of the field layout such as touch and goal lines, goal and penalty area, center line and mark, corner and penalty arcs, center mark and center circle, and penalty marks. A complete, continuous and smooth path is then generated by connecting these waypoints using 2D Dubins’ path in a way to ensure that the generated path takes into account the dynamic constraints of the vehicle (such as maximum curvature and velocity), keep the vehicle at a safe distance from obstacles, and not harm the field grass. The efficiency of the algorithm is tested using simulation and in reality. Results showed that the algorithm is able to reliably plan a safe path in real time able to command the line marking robot with high accuracy and without the need for human guidance. The path planning algorithm developed in this paper is implemented in both Matlab and Python.