Motion Planning Under Temporal Logic Specifications In Semantically Unknown Environments

2025-11-05

AI Summary

This paper addresses robust spatiotemporal motion planning under syntactically co-safe linear temporal logic (scLTLₙₑₓₜ) specifications in environments where semantic classes are unknown, their spatial locations uncertain, yet prior probabilities over semantic labels are available. We propose a novel method integrating probabilistic modeling with automata theory: (i) constructing a semantics-aware special product automaton that explicitly encodes environmental uncertainty into its structure; (ii) designing edge-dependent reward functions to enable online replanning; and (iii) combining value iteration with real-time decision-making for closed-loop planning. We formally prove correctness—i.e., guaranteed task satisfaction—and convergence of the planning algorithm. Extensive simulations and physical experiments demonstrate the method’s effectiveness, adaptability to dynamic uncertainties, and computational feasibility in realistic settings.

Technology Category

• Planning, Routing, and Scheduling: Planning under Uncertainty
• Intelligent Robots: Motion and Path Planning
• Multiagent Systems: Multiagent Planning

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

This paper addresses a motion planning problem to achieve spatio-temporal-logical tasks, expressed by syntactically co-safe linear temporal logic specifications (scLTL ext), in uncertain environments. Here, the uncertainty is modeled as some probabilistic knowledge on the semantic labels of the environment. For example, the task is"first go to region 1, then go to region 2"; however, the exact locations of regions 1 and 2 are not known a priori, instead a probabilistic belief is available. We propose a novel automata-theoretic approach, where a special product automaton is constructed to capture the uncertainty related to semantic labels, and a reward function is designed for each edge of this product automaton. The proposed algorithm utilizes value iteration for online replanning. We show some theoretical results and present some simulations/experiments to demonstrate the efficacy of the proposed approach.

Problem

• Motion planning under temporal logic specifications in uncertain environments
• Achieving spatio-temporal tasks with probabilistic semantic label knowledge
• Online replanning for tasks with unknown region locations using value iteration

Innovation

• Automata-theoretic approach with product automaton
• Value iteration for online replanning
• Reward function designed for product automaton edges