Factorizing Diffusion Policies for Observation Modality Prioritization

摘要

Diffusion models have been extensively leveraged for learning robot skills from demonstrations. These policies are conditioned on several observational modalities such as proprioception, vision and tactile. However, observational modalities have varying levels of influence for different tasks that diffusion polices fail to capture. In this work, we propose 'Factorized Diffusion Policies' abbreviated as FDP, a novel policy formulation that enables observational modalities to have differing influence on the action diffusion process by design. This results in learning policies where certain observations modalities can be prioritized over the others such as $\texttt{vision>tactile}$ or $\texttt{proprioception>vision}$. FDP achieves modality prioritization by factorizing the observational conditioning for diffusion process, resulting in more performant and robust policies. Our factored approach shows strong performance improvements in low-data regimes with $15\%$ absolute improvement in success rate on several simulated benchmarks when compared to a standard diffusion policy that jointly conditions on all input modalities. Moreover, our benchmark and real-world experiments show that factored policies are naturally more robust with $40\%$ higher absolute success rate across several visuomotor tasks under distribution shifts such as visual distractors or camera occlusions, where existing diffusion policies fail catastrophically. FDP thus offers a safer and more robust alternative to standard diffusion policies for real-world deployment. Videos are available at https://fdp-policy.github.io/fdp-policy/ .