Co-Evolving Robot Orchestrators and Policies through Deployment

Key points

• Vision-language-action (VLA) policies trained on large datasets are capable within their training domains, yet they still fail to generalize to the variety of situations a robot meets in real-world deployment.1
• Agentic robot systems complement the policy with a vision-language model (VLM) orchestrator that learns when to call the policy, how to instruct it, and when to use scripted skills instead.1
• Across ten simulated RoboLab tasks, Robo-COP raises mean held-out success from 64.8% to 73.8% over the same harness with a frozen policy, while fine-tuning on a fixed schedule without verification reaches only 65.8%.1
• Robo-COP turns deployment into a self-improving flywheel in which robots learn by doing, with each improvement in execution producing better data for the next round of learning.1

Sources (1)

• [1]Co-Evolving Robot Orchestrators and Policies through Deployment
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 11:44 PM

  We propose Robo-COP, in which the orchestrator and policy co-evolve during deployment.

  Vision-language-action (VLA) policies trained on large datasets are capable within their training domains, yet they still fail to generalize to the variety of situations a robot meets in real-world deployment.

Extractive summary: sentences quoted from the sources.