A random selection fuzzy rule base for robust target approaching behavior of a mobile robot

In this paper we present a simple control architecture of a mobile robot for the purpose of target approaching. The proposed architecture combines three separate control modules to achieve an overall robot behavior for robust target approaching in a constrained environment. The three modules are "target search", "target approach" and "fuzzy obstacle avoidance". The target search module applies a constant angular velocity to search for the target if is out of the sensing range of the robot. The target approach module generates the linear and angular velocities to navigate the mobile robot on a path straight towards the target. The fuzzy obstacle avoidance contains our proposed random selection fuzzy rule base (RS_FRB) to overcome the problem of local trapping of the mobile robot in continuous wandering loops and for sensor based obstacle avoidance. The three control modules are fused together to generate a linear and angular velocity for the mobile robot. The result is an overall control behavior that is able to navigate the mobile robot towards a target while avoiding any obstacles in its path. Simulation results show the effectiveness of the proposed approach in reaching the target position even if the mobile robot is trapped inside a U-shaped obstacle.