Perception-Based Temporal Logic Planning in Uncertain Semantic Maps

TL;DR

This paper presents a novel perception-based temporal logic planning framework for multi-robot systems operating in uncertain semantic maps, enabling adaptive control policies to accomplish complex missions under environmental and perceptual uncertainties.

Contribution

It introduces a new sampling-based planning algorithm that generates and updates control policies for robots based on perception-driven temporal logic specifications in uncertain environments.

Findings

Efficient planning architecture demonstrated through extensive experiments.

Adaptive control policies effectively handle environmental and perceptual uncertainties.

Framework successfully accomplishes complex missions with probabilistic satisfaction requirements.

Abstract

This paper addresses a multi-robot planning problem in environments with partially unknown semantics. The environment is assumed to have known geometric structure (e.g., walls) and to be occupied by static labeled landmarks with uncertain positions and classes. This modeling approach gives rise to an uncertain semantic map generated by semantic SLAM algorithms. Our goal is to design control policies for robots equipped with noisy perception systems so that they can accomplish collaborative tasks captured by global temporal logic specifications. To specify missions that account for environmental and perceptual uncertainty, we employ a fragment of Linear Temporal Logic (LTL), called co-safe LTL, defined over perception-based atomic predicates modeling probabilistic satisfaction requirements. The perception-based LTL planning problem gives rise to an optimal control problem, solved by a novel sampling-based algorithm, that generates open-loop control policies that are updated online to adapt to a continuously learned semantic map. We provide extensive experiments to demonstrate the efficiency of the proposed planning architecture.