Path Planning of Mobile Robot Based on Modification of Vector Field Histogram using Neuro-Fuzzy Algorithm

In this paper, a neuro-fuzzy algorithm has been implemented to improve the path planning of a mobile robot based on modification of vector field histogram (VFH) approach using neuro-fuzzy algorithm. A back propagation neural network has been used for detection of unknown obstacle to avoid collision. A neural network model is used to learn (off line) many critical status of obstacle topology. Fuzzy rules were applied to steer the mobile robot to the target direction. The candidate valleys generated by neural network from local sensor data in the robot would be fuzzified to select a winning valley that is nearest to the target direction. The simulation result of the proposed algorithm shows good navigation properties and ability to overcome the limitation of traditional vector field histogram approach. This work was implemented using Matlab® and Mobotsim®