Determination of Minimum-Time Velocity Profile for Trajectory Planning of Small Size League Robots

• Lucas G. Neves ITA
• Guilherme Müller Bertolino ITA
• Marcos R. O. A. Maximo ITA
• Gabriela Werner Gabriel ITA

Resumo

This article proposes using the Minimum-Time Velocity Profile (MTVP) algorithm in the Small Size League (SSL) of RoboCup. This algorithm solves the trajectory planning problem, providing insight on how to navigate an already planned path while accounting for dynamic and kinematic constraints. In this work, the path is composed by line segments and circular arcs, which is generated by the team's visibility-graph-based path planner. However, this is not a limitation of the developed algorithm, which remains applicable to more general paths. Experimental results demonstrate the effectiveness of the MTVP approach: the variant that considers angular acceleration constraints achieved a 1.1% improvement in navigation time over a baseline P controller with saturations, while the variant that ignores this constraint achieved an 8.57% improvement. This work contributes with a solution to the velocity profile determination for trajectory planning in the SSL environment.

Palavras-chave: Monte Carlo methods, Trajectory planning, Navigation, Heuristic algorithms, Education, Kinematics, Motors, Trajectory, Robots, Faces, Trajectory Planning, Minimum-Time Velocity Profile, Small Size League