What problem does it solve? Autonomous robots using AMCL can lose localization due to kidnapping, odometry spikes, symmetric environments, or map changes, causing navigation failures and collisions. This Skill provides systematic detection symptoms, root-cause analysis, and concrete recovery procedures to restore accurate pose estimation. ## Core Features & Use Cases - Failure Diagnosis: Identify localization loss from observable symptoms like erratic RViz jumps, costmap mismatches, TF timeouts, and rising AMCL covariance. - Three Recovery Strategies: Global re-localization via the /reinitialize_global_localization service, manual initial pose via RViz or the /initialpose topic, and navigating to geometrically distinctive landmarks. - Automated Monitoring: A Python covariance-monitoring node that watches /amcl_pose and triggers global re-localization when variance exceeds a threshold, plus Behavior Tree integration with spin-and-wait recovery subtrees. - Use Case: A warehouse robot's AMCL particles diverge after a wheel slip; the monitor node detects covariance above 0.5, calls global re-localization, spins the robot to gather scans, and navigation resumes once the filter reconverges. ## Quick Start Ask the AI to diagnose why the robot's position jumps erratically in RViz and set up automated AMCL covariance monitoring with global re-localization recovery.
