In the paper the behavior based control system of an autonomous mobile robot SCORPION is presented to execute the one of the most difficult navigation task, which is the complete coverage task of unknown area with static obstacles in the environment. The main principle assumed to design control system was that the robot should cover all area only once, if it possible, to optimize the length of path and energy consumption. All commercial robots like Roomba, Trilobite or IVO move using structured templates combined with random movement. Therefore the path of coverage is not optimal directions of movement are often chosen randomly, so robot covers the same area many times wasting time and energy. In paper the five main developed templates of movement were described to fulfill main task in ordered manner using primarily the way of the ox template of coverage [1, 2, 5, 1. The behavioral control system is implemented in a computer application written in Python [5]. In the paper the test methodology of the developed system on real mobile robot ERSP SCORPION equipped with IR sensors is presented. Graphical and quantitative results of tests of accomplishment of complete coverage task are given for 6 different configurations of obstacles in the robots environment. Conclusions are presented and discussed [5]. Ways to improve the quality indicators [1, of the task of complete coverage of a unknown area are also showed.
Q-Learning for Autonomous Mobile Robot Obstacle Avoidance