Open Access Open Access  Restricted Access Subscription Access

Acquisition of Multiple Moving Obstacles Avoidance Actions for Wheeled Type Robots

Tomohiro Yamaguchi, Yoshio Watanabe


Mobile robot path planning in a movement environment is an important problem. We studied acquisition of a path to a destination and obstacle avoidance of a wheeled type robot. The paper proposes a method of path planning based on neural network and genetic algorithm. The avoidance action of a wheeled type robot is determined from the obstacle configuration, the robot’s self-state and destination information using a neural network. The design parameter of neural network is adjusted by using genetic algorithm. The effectiveness of present method is proved through a simulation.

Full Text:



J. Miura and Y. Shirai, “Probabilistic uncertainty modeling of obstacle motion for robot motion planning,” Journal of Robotics and Mechatronics, Vol.14, No.4, pp. 349–356, 2002.

P. Fiorini and Z. Shiller, “Motion planning in dynamic environments using velocity obstacles,” International Journal of Robotics Research, Vol.17, No.7, pp. 760–772, 1998.

A. Inoue, K. Inoue and Y. Okawa, “On-line motion planning of an autonomous mobile robot to avoid multiple moving obstacles based on the prediction of their future trajectories,” Journal of Robotics Society of Japan,

Vol.15, No.2, pp. 249–260, 1997.

Elanayar VTS, Shin YC, “Radial basis function neural network for approximation and estimation of nonlinear stochastic,” IEEE Transactions on Neural Networks, Vol.5, No.4, pp. 594–603, 1994

Z. Michalewicz, Genetic Algorithms + Data Structure = Evolution Programs, 3rd, revised and extended edition ed., Springer, Germany, 1996



  • There are currently no refbacks.