🤖 AI Summary
This work addresses the failure of policy transfer from simulation to reality caused by discrepancies in visual rendering and physical dynamics. To bridge this sim-to-real gap, the authors propose a cross-domain dual imitation learning approach that leverages a shared history encoder to jointly model domain-invariant features across visual and dynamical domains in observation space. By mapping observation–action sequences that yield equivalent long-term behaviors to nearby latent states, the method enables zero-shot sim-to-real policy transfer without requiring separate adaptation modules. Evaluated on visually guided navigation, contact-rich manipulation, and visual servoing tasks, the approach substantially outperforms existing domain adaptation and co-training baselines, achieving the first end-to-end, highly effective zero-shot sim-to-real transfer.
📝 Abstract
Sim2real transfer for robot policy learning suffers due to mismatch between simulation and reality. Existing methods typically address each gap in isolation through separate adaptation modules, which are composed or layered when both gaps coexist. Yet the basis for attempting sim2real in the first place is that there is shared structure between a task in simulation and reality, where equivalent actions from equivalent configurations produce equivalent long term outcomes regardless of domain specific differences in rendering or physics. In this paper, we study whether we can identify and exploit this shared structure from raw observations to train a policy that enables zero shot transfer. We introduce BIFROST, which learns a shared history encoder on paired cross-domain data via cross-domain bisimulation objective: observation-action sequences leading to equivalent long-term behavior are mapped to nearby latent states, regardless of domain. Policies trained on these latent states in simulation transfer zero-shot to reality. We provide empirical evidence on sim2sim visual navigation and sim2real contact rich manipulation task and visual servoing task that BIFROST achieves effective transfer where domain adaptation and co-training baselines fail under both visual and dynamics domain gaps.