Motion planning simulator for holonomic robot soccer platform

Abstract

This paper presents a soccer robot simulator which is built base on holonomic wheeled soccer robot platform. The platform is self-made with specifications based on Middle Sized League RoboCup rules. The simulator is built for multi-operating systems that can be run on Windows, Mac or Linux operating system. The use of simulators can speed up the development of algorithms for wheeled soccer robots because algorithmic testing can be done in the simulator before being implemented in actual robots. Time and cost saving are the huge benefits that can be derived from using this simulator. Algorithm A∗ is used for obstacle avoidance testing. The result shows that algorithm programs written in the simulator can be directly implemented into a real soccer robot platform. This simulator has motion planning RMSE equal to 6.5 cm for path planning without obstacle avoidance and RMSE equal to 46.6 cm for path planning with obstacle avoidance algorithm.