This project was created for 16-833: Simultaneous Localization and Mapping (Prof. Michael Kaess).
Collaborators: Sandra Sajeev
Monte Carlo Localization is a technique that uses many “particles” initialized on a map. These particles move with Gaussian noise according to the odometry observed by the robot, then are resampled using importance sampling to better fit the measurements the robot observes. Particles are then transformed to more likely locations after resampling and the process is repeated until convergence.
C++
OpenCV
Eigen
OpenMP