SOLUTION FOR TRAP SITUATIONS OF WHEELED MOBILE ROBOT DUE TO LOCAL MINIMA PROBLEMS

This paper presents a novel scheme for the mobile robot motion planning and navigation. Both global and local motion planning have been considered by implementing the planned scheme and the behavior based approach. For global motion planning, planned scheme has been implemented by using the a priori knowledge of the environment. For local motion planning the path is divided into sub-paths. Behavior based approach has been implemented for local motion planning. For every sub-goal the planned path is updated by using the sensors data of the environment of the robot. Three schemes of Obstacle avoidance, Follow the wall and Path following have been implemented. A novel technique has been proposed which autonomously change the weight of the appropriate scheme. Through fuzzy inference system the weight of the required scheme among the three is increased depending upon the situations. Fuzzy rules have been formed for safe navigation of robot to ensure that the robot must reach the target without any collision with obstacles. Depending upon the particular situations, the weight of a suitable scheme among obstacle avoidance, follow the wall and path following is updated. This weight updating of the appropriate scheme for specific situation eliminates trap situation of the mobile robot produced by the local minima problems. For validation of the proposed technique simulation results have been presented