Provably Safe Stein Variational Clarity-Aware Informative Planning

2025-11-13

AI Summary

Autonomous robots performing information-gathering tasks in dynamic environments face dual challenges: inaccurate spatiotemporal modeling of information decay and difficulty in rigorously enforcing safety constraints. To address these, we propose a clarity-aware informative trajectory planning framework. First, we formulate a spatially heterogeneous information decay model to capture location-dependent environmental uncertainty. Second, we embed this non-uniform decay mechanism into a trajectory optimization framework, jointly optimizing the information-motion distribution via Stein variational inference and enhancing robustness of information measurement through differential entropy normalization. Third, we introduce a gate-based safety filtering layer that provides provably safe low-level control. Extensive simulations and real-robot experiments demonstrate that our method significantly reduces information loss rate while maintaining persistent safe navigation, thereby improving long-term perception quality.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Intelligent Robots: State Estimation
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Security and Privacy: Data transparency and provenance
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Autonomous robots are increasingly deployed for information-gathering tasks in environments that vary across space and time. Planning informative and safe trajectories in such settings is challenging because information decays when regions are not revisited. Most existing planners model information as static or uniformly decaying, ignoring environments where the decay rate varies spatially; those that model non-uniform decay often overlook how it evolves along the robot's motion, and almost all treat safety as a soft penalty. In this paper, we address these challenges. We model uncertainty in the environment using clarity, a normalized representation of differential entropy from our earlier work that captures how information improves through new measurements and decays over time when regions are not revisited. Building on this, we present Stein Variational Clarity-Aware Informative Planning, a framework that embeds clarity dynamics within trajectory optimization and enforces safety through a low-level filtering mechanism based on our earlier gatekeeper framework for safety verification. The planner performs Bayesian inference-based learning via Stein variational inference, refining a distribution over informative trajectories while filtering each nominal Stein informative trajectory to ensure safety. Hardware experiments and simulations across environments with varying decay rates and obstacles demonstrate consistent safety and reduced information deficits.

Problem

• Planning safe informative paths for robots in dynamic environments
• Modeling spatially varying information decay in robotic exploration
• Ensuring safety through trajectory optimization and filtering mechanisms

Innovation

• Stein variational inference for trajectory distribution learning
• Clarity dynamics embedded in trajectory optimization
• Low-level filtering mechanism for provable safety enforcement