Application of Improved Back Propagation Neural Network in Mowing Robot’s Path Planning

• Published in: Applied Mechanics and Materials Vol. 602-605; no. Advanced Manufacturing and Information Engineering, Intelligent Instrumentation and Industry Development; pp. 916 - 919
• Main Authors: Yang, Guo Wei, Cui, Xue Mei, Wu, Shao Long, Deng, Yan
• Format: Journal Article
• Language: English
• Published: Zurich Trans Tech Publications Ltd 01.08.2014
• Subjects: Algorithms; Back propagation; Computer simulation; Grid method; Mowing; Neural networks; Path planning; Robots; BP Neural Network (BPNN); Mowing Robot; Path Planning; Complete Coverage
• ISBN: 9783038351948; 3038351946
• ISSN: 1660-9336; 1662-7482; 1662-7482
• DOI: 10.4028/www.scientific.net/AMM.602-605.916

To address the problem of planning complete coverage paths for mowing robots that have the greatest coverage rates and the lowest repetitive rates, we proposed an improved back propagation neural network algorithm based on priority traversal thoughts for local path planning. The algorithm based on plowing global path planning. We adopted grid method to model the environment and used Matlab2010a to simulate for the algorithm. Simulation results show that the proposed algorithm can make the mowing robot walk out of dead zone, the dead zone was composed of obstacle grid or the grid that had been cut around this area, and achieve the complete coverage path planning.