Fuzzy logic based control of mobile robot navigation: A case study on iRobot Roomba Platform

This work addresses the issue of mobile robot navigation system. In this study, the proposed navigation system is introduced. It combines a deliberative planning and a reactive/behaviour based control, based upon the motor schema technique into hybrid control architecture. Behaviour based robots have successfully demonstrated their performance in reacting to different working environments. However, it is difficult to coordinate different robot behaviours in order to achieve a specific task. To overcome this difficulty, a behaviour coordination module has been developed using weighted vector summation of the schemas. A set of behaviour weights employs both the fixed priorities and dynamic weights. The adaptation of dynamic weights depends on the current environment condition, and is achieved through a fuzzy inference system where fuzzy rules are used to automatically generate and update the weights during navigation. This allows the robot to successfully complete its task in a robust, smooth, and speedy manner without introducing any serious problems. A series of experiments has been conducted and the results revealed that the proposed navigation system can safely and effectively navigate the mobile robot around a designated environment.   Key words: Behaviour co-ordination, fuzzy logic control, motor schema, roomba robot platform.