Optimization of Rao Blackwellized Particle Filter SLAM using Firefly algorithm

  • Muhammad Fayyaz University of Engineering and Technology Peshawar
  • Muhammad Tufail Department of Mechatronics Engineering, University of Engineering & Technology Peshawar, Pakistan
  • Shahzad Anwar Department of Mechatronics Engineering, University of Engineering & Technology Peshawar, Pakistan
  • Zubair Ahmad Department of Mechatronics Engineering, University of Engineering & Technology Peshawar, Pakistan
  • Shahbaz Khan Department of Mechatronics Engineering, University of Engineering & Technology Peshawar, Pakistan
Keywords: Machine learning, Mobile robot navigation, Optimization

Abstract

Navigation accuracy, which is an imperative performance indicator for mobile robots, is intimately associated with the grid mapping algorithm (G-mapping) accuracy. In an unstructured environment, mobile robot positioning accuracy is important to ensure safety. For this reason, in this study G-mapping Algorithm is modelled based on Rao-Blackwellized particle filter (RBPF) offering better results with a low number of sensors and features. To investigate various methods' effectiveness, a comparative analysis of three optimization methods namely Gradient descent, ANT colony, and firefly algorithm was made. The results exhibit that the firefly method performs well in terms of navigation accuracy, particle degradation, and ensuring mobile robot safety in a complex and unstructured environment.

Published
2020-12-14
How to Cite
[1]
M. Fayyaz, M. Tufail, S. Anwar, Z. Ahmad, and S. Khan, “Optimization of Rao Blackwellized Particle Filter SLAM using Firefly algorithm”, PakJET, vol. 3, no. 03, pp. 46-50, Dec. 2020.