A new CNN-Based method of path planning in dynamic environment

  • Authors:

  • Maciej Przybylski;Barbara Siemiątkowska

  • Affiliations:

  • Institute of Automatic Control and Robotics, Warsaw University of Technology, Warsaw, Poland;Institute of Automatic Control and Robotics, Warsaw University of Technology, Warsaw, Poland

  • Venue:
  • ICAISC'12 Proceedings of the 11th international conference on Artificial Intelligence and Soft Computing - Volume Part II
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper the problem of path planning in a dynamic environment is considered. The minimum cost path is planned using Cellular Neural Network (CNN). CNN is a very useful tool for parallel signal processing and can be implemented using VLSI. In the proposed approach the problem of local minima (dead ends on a map) does not exist. Different criteria can be taken into account during path planning for example: the size of the robot, the traversability cost, the occurrence of dynamic obstacles, etc. The method allows us to specify the goal using semantic labels. The experiments performed in a real static environment and simulations in a dynamic environment have shown the efficiency of the proposed method.