Newest papers

Simulation Priors for Rapid Adaptation in Model-Based RL for Autonomous Racing

Adrian Remonda, Jiajun Xi, Nicklas Hansen, Marcus Greiff, Thomas Lew, John Talbot, John K. Subosits, Xiaolong Wang · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26757 · cited by 0 works (OpenAlex) · open-access PDF

Korean

강화 학습은 고성능 제어 정책을 배울 수 있지만 광범위한 상호 작용에 대한 의존성은 데이터 수집이 비싸고 실패가 안전하지 않은 물리적 시스템에 배포를 제한합니다. 우리는 미리 알려진 적응 (PIA)을 도입합니다. 오프라인-온라인 모델 기반 RL 프레임 워크프레임은 빠른 적응을 위해 구조화 된 시뮬레이션 프리어를 사용합니다. PIA는 차별화 된 낮은 신뢰성과 높은 신뢰성의 시뮬레이션 경로에 세계 모델을 사전 훈련시킵니다. 우리는 불확실성 규칙화 된 계획을 사용하여 제한된 타겟 도메인 상호 작용으로 그것을 조정합니다. 정책 전송을위한 정확한 시뮬레이터를 요구하는 대신 PIA는 시

Machine translation (Meta M2M100); the English original follows.

Abstract

Reinforcement learning enables learning high-performance control policies, but its reliance on extensive interaction limits deployment on physical systems where data collection is costly, and failures are unsafe. We introduce Prior-Informed Adaptation (PIA), an offline-to-online model-based RL framework that uses structured simulation priors for rapid adaptation. PIA pretrains a world model on heterogeneous low- and high-fidelity simulated trajectories, then adapts it with limited target-domain interaction using uncertainty-regularized planning. Rather than requiring an accurate simulator for policy transfer, PIA treats simulated trajectories as biased priors whose residual mismatch is corrected during online adaptation. We evaluate PIA in autonomous racing using high-fidelity simulation and on a real full-size autonomous race car. In simulation, PIA adapts faster across track and surface shifts than strong baselines, achieving lower lap times and fewer failures. On hardware, PIA achieves reliable few-shot adaptation with less than one hour of on-vehicle interaction.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Prior probability, Reinforcement learning, Computer science, Adaptation (eye), Artificial intelligence, Software deployment

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange