DynaHarness: A Dynamic Physical Harness for Self-Evolving Robot Agents

Key points

Instead of treating a failure as 'the robot was bad', it localized which part (planning, execution, or recovery) went wrong and fixed only that part.
It split the slow brain (planning) and the fast brain (monitoring) into two layers on different time scales, so ambiguous commands are refused and alternative actions are sought.
Before a fix is admitted, it must pass regression checks, preventing the risk of breaking previously working functions while repairing.
On 800 new LIBERO-Pro initial states it scored 75.2% versus 17.5% for the frozen policy, more than quadrupling performance with the harness alone and no retraining.

External research summaries. These are not HDATF publications or measured product results.