Risk-Guided Diffusion: Toward Deploying Robot Foundation Models in Space, Where Failure Is Not An Option
This addresses the critical need for reliable robotic navigation in space exploration missions where failures are unacceptable, representing a domain-specific incremental improvement.
The paper tackles the problem of safe navigation for space robots in extreme terrain by proposing a risk-guided diffusion framework that combines fast learning with slow physics-based reasoning. Their approach reduces failure rates by up to 4× while maintaining goal-reaching performance, as demonstrated in hardware experiments at a Mars-analog facility.
Safe, reliable navigation in extreme, unfamiliar terrain is required for future robotic space exploration missions. Recent generative-AI methods learn semantically aware navigation policies from large, cross-embodiment datasets, but offer limited safety guarantees. Inspired by human cognitive science, we propose a risk-guided diffusion framework that fuses a fast, learned "System-1" with a slow, physics-based "System-2", sharing computation at both training and inference to couple adaptability with formal safety. Hardware experiments conducted at the NASA JPL's Mars-analog facility, Mars Yard, show that our approach reduces failure rates by up to $4\times$ while matching the goal-reaching performance of learning-based robotic models by leveraging inference-time compute without any additional training.