Browse Teach Curate

Adaptive Monte Carlo Localization of Mobile Robots in Known Maps

A particle-filter approach to estimating a mobile robot's position and orientation within an already-known map by maintaining many weighted hypotheses (particles) that are moved and re-weighted as the robot acts and senses. The adaptive variant dynamically varies the number of particles over time so that many are used when uncertainty is high and few when the pose has converged, reducing computation while handling global localization and recovery.

KITT Text

Robot localization in a mapped environment using Adaptive Monte Carlo algorithm Sagarnil Das

This paper describes implementing and testing Adaptive Monte Carlo Localization (AMCL), a particle-filter method, to estimate a mobile robot's pose within a previously mapped environment and drive it…