Sim-Anchored Learning for On-the-Fly Adaptation

TL;DR

This paper introduces a multi-objective optimization approach using anchor critics to enable real-time adaptation of RL agents, preserving important behaviors learned in simulation while adapting to real-world data, demonstrated on robotics tasks.

Contribution

It proposes a novel multi-objective framework with anchor critics for on-the-fly RL adaptation, addressing catastrophic forgetting and behavior preservation.

Findings

Robust behavior retention in sim-to-sim benchmarks.

Successful sim-to-real adaptation with power savings in quadrotor.

Open-source SwaNNFlight firmware for live adaptation.

Abstract

Fine-tuning simulation-trained RL agents with real-world data often degrades crucial behaviors due to limited or skewed data distributions. We argue that designer priorities exist not just in reward functions, but also in simulation design choices like task selection and state initialization. When adapting to real-world data, agents can experience catastrophic forgetting in important but underrepresented scenarios. We propose framing live-adaptation as a multi-objective optimization problem, where policy objectives must be satisfied both in simulation and reality. Our approach leverages critics from simulation as "anchors for design intent" (anchor critics). By jointly optimizing policies against both anchor critics and critics trained on real-world experience, our method enables adaptation while preserving prioritized behaviors from simulation. Evaluations demonstrate robust behavior retention in sim-to-sim benchmarks and a sim-to-real scenario with a racing quadrotor, allowing for power consumption reductions of up to 50% without control loss. We also contribute SwaNNFlight, an open-source firmware for enabling live adaptation on similar robotic platforms.