SLAMSanity

SLAMSanity is a desktop app (with optional web dashboard) that automatically stress-tests SLAM pipelines on your recorded bags and sim logs, then flags the exact segments where tracking degrades. You import ROS/ROS2 bag files (or common log formats), choose your SLAM stack profile (e.g., ORB-SLAM3, Cartographer, RTAB-Map, VINS-Fusion), and the app runs a battery of repeatable checks: loop consistency, drift rate, relocalization frequency, map entropy, and sensor timestamp sanity. It outputs a concise “map health report” with charts, annotated timelines, and exportable JSON for CI. It’s a combination traditional + AI app: traditional metrics plus an AI-assisted root-cause assistant that correlates failures with motion patterns, lighting changes, feature scarcity, and sensor anomalies. This is not a magic fix; it’s a pragmatic QA gate for SLAM reliability.