/Monte-Carlo-Localization

A Monte Carlo or Particle Filter approach for robot localization.

Primary LanguagePython

Monte Carlo Localization

  • This is a Python implementation of the Monte Carlo Localization algorithm for robot movement data obtained by a turtle-bot within a university classroom (CSE_668.bag).
  • Work done as part of CSE 668 - Advanced Robotics taught by Nils Napp at the University at Buffalo.
  • landmarks.csv contains landmark location data.
  • Uses rosbag to read the BAG file, hence ROS needs to be installed. Tested on ROS Kinetic.
  • Run the code:
python monte-carlo.py --steps 20 --num_particles 200
  • steps - Number of localization steps for the Monte Carlo algorithm.
  • num_particles - Number of particles to initialize the particle filter with.