Abstract—This paper introduces a new path-motion planning method for autonomous mobile robot which should move safely in unknown Dynamic environment. The environment may have numbers of obstacles of arbitrary shape and obstacles are allowed to move. We describe our approach to solve the motion-planning problem, using neural networks-based technique .The neural network is used to model the obstacles motion and predict their movements. According to one step ahead prediction of obstacle’s trajectory, a rule-based technique for avoiding obstacles by adjusting the robot’s velocity is suggested .This sensor based online method gives a safe but near optimal path. Simulation examples of generated path with proposed techniques will be presented and show the efficiency of the method.
Index Terms—Mobile robot, dynamic environment, neural network, path planning, speed control.
Soheila Dadelahi is with the Young Researcher’s Club, Buin Zahra Branch, Islamic Azad University, Buin Zahra Iran (e-mail: dadelahi_61@yahoo.com)
Mohammad Reza Jahed Motlagh and Ali Reza Mohammad Shahri are with the Iran University of Science and Technology (e-mail: jahedmr@iust.ac.ir, Shahri@iust.ac.ir)
Cite: Soheila Dadelahi, Mohammad Reza Jahed Motlagh, and Ali Reza Mohammad Shahri, "Obstacle Displacement Prediction for Robot Motion Planning and Velocity Changes," International Journal of Information and Electronics Engineering vol. 3, no. 3, pp. 270-274, 2013.