Embodied Robotics: Planning, Navigation, and Recovery

What to do

1. Test planning as a sequence of decisions.
2. Test maps, localization, obstacle avoidance, replanning, elevators, doors, ramps, reflective surfaces, narrow spaces, crowds, and no-go zones.
3. Measure path length, time, blocked-path handling, human proximity, speed, near misses, and how quickly the system recognizes that the route is no longer valid.
4. Do not stop at asking whether the robot noticed a low-confidence perception, blocked path, lost connection, power problem, unsafe command, or failing actuator.
5. Measure warning quality, time to risk reduction, safe placement of objects, controlled stopping distance, fallback navigation, human handoff, state preservation, and behavior when the first recovery strategy also fails.

Evidence to preserve

• Record the full action trace so failures can be replayed and scored by trajectory, not remembered as folklore.
• Preserve the inputs, versions, configurations, raw outcomes, and results for embodied robotics planning navigation recovery needed to reproduce work on Embodied Robotics: Planning, Navigation, and Recovery.
• Report results for embodied robotics planning navigation recovery by relevant slice, separate blocker failures from averages, state uncertainty and blind spots, and connect the result to a release decision.