Abstract
In this paper, a new intelligent motion planning approach to mobile robot navigation is addressed. Adaptive neuro-fuzzy inference system (ANFIS) is a well-known hybrid neuro-fuzzy structure for modelling the engineering system. It has also taken the advantages of both learning ability of neural network and the reasoning ability of the fuzzy inference system. In this navigational model, different sensor-extracted information, such as front obstacle distance, right obstacle distance, left obstacle distance, heading angle, left wheel velocity and right wheel velocity, are given input to the ANFIS controller and output from the controller is steering angle for the robot. Based on the output information, the robot moves safely in an unstructured environment populated by variety of static obstacles. Using ANFIS tool box, the obtained mean of squared error for the training dataset in the current paper is 0.0021. Finally, the simulation results are verified with real-time experimental results using Khepra-III mobile robot to show the feasibility and effectiveness of the proposed navigational algorithm.
Disclosure statement
No potential conflict of interest was reported by the authors.