Risk-Averse Traversal of Graphs with Stochastic and Correlated Edge Costs for Safe Global Planetary Mobility

2025-05-19

AI Summary

This paper addresses risk-sensitive global path planning for long-distance autonomous navigation on planetary surfaces (e.g., Mars) under uncertain, correlated terrain conditions, where edge costs in the graph model are stochastic and mutually dependent, and the objective is to minimize tail-risk exposure via Conditional Value-at-Risk (CVaR). Method: We pioneer the integration of CVaR into the Canadian Traveller Problem framework, introduce a novel stochastic graph representation capable of encoding cost correlations, and design the first exact AND-OR graph search algorithm for computing CVaR-optimal policies. Contribution/Results: Extensive simulations using real Mars orbital and topographic data demonstrate that the method generates adaptive, risk-averse traversal strategies; information-guided detours significantly reduce overall traversal risk, yielding substantially improved robustness over conventional expected-cost optimization.

Technology Category

• Planning, Routing, and Scheduling: Planning under Uncertainty
• Search and Optimization: Heuristic Search
• Intelligent Robots: Motion and Path Planning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

In robotic planetary surface exploration, strategic mobility planning is an important task that involves finding candidate long-distance routes on orbital maps and identifying segments with uncertain traversability. Then, expert human operators establish safe, adaptive traverse plans based on the actual navigation difficulties encountered in these uncertain areas. In this paper, we formalize this challenge as a new, risk-averse variant of the Canadian Traveller Problem (CTP) tailored to global planetary mobility. The objective is to find a traverse policy minimizing a conditional value-at-risk (CVaR) criterion, which is a risk measure with an intuitive interpretation. We propose a novel search algorithm that finds exact CVaR-optimal policies. Our approach leverages well-established optimal AND-OR search techniques intended for (risk-agnostic) expectation minimization and extends these methods to the risk-averse domain. We validate our approach through simulated long-distance planetary surface traverses; we employ real orbital maps of the Martian surface to construct problem instances and use terrain maps to express traversal probabilities in uncertain regions. Our results illustrate different adaptive decision-making schemes depending on the level of risk aversion. Additionally, our problem setup allows accounting for traversability correlations between similar areas of the environment. In such a case, we empirically demonstrate how information-seeking detours can mitigate risk.

Problem

• Develops risk-averse path planning for planetary robots with uncertain edge costs
• Solves CVaR-optimal policies for safe traversal in stochastic correlated environments
• Validates approach using Martian terrain maps and adaptive decision-making

Innovation

• Risk-averse variant of Canadian Traveller Problem
• Exact CVaR-optimal policies via AND-OR search
• Traversability correlations for risk mitigation