Search papers, labs, and topics across Lattice.
This paper investigates the impact of risk-aware domain randomization (DR) on contact-rich sampling-based predictive control (SPC) using a Push-T task. They compare average, optimistic, and pessimistic rollout aggregations under randomized model instances, revealing that DR significantly alters the effective cost landscape. The key finding is that DR reshapes the basin of attraction around contact-producing actions, influencing the optimizer's behavior beyond simply improving robustness.
Domain randomization doesn't just make your robot policies more robust; it fundamentally warps the optimization landscape, potentially guiding your search towards better contact-rich behaviors.
Domain randomization (DR) is widely used in policy learning to improve robustness to modeling error, but remains underexplored in contact-rich sampling-based predictive control (SPC), where rollout quality is highly sensitive to uncertainty. In this work, we take the first step by studying risk-aware DR in predictive sampling on a simple yet representative Push-T task, comparing average, optimistic, and pessimistic rollout aggregations under randomized model instances. Our initial results suggest that DR affects not only robustness to model error, but also the effective cost landscape seen by the sampling-based optimizer, by reshaping the basin of attraction around contact-producing actions. This opens up potential for exploring better grounded risk-aware contact-rich SPC under model uncertainty. Video: https://youtu.be/f1F0ALXxhSM
