Motion Planning Under Temporal Logic Specifications in Semantically Unknown Environments

Published in American Control Conference (ACC) 2026, 2026

This paper addresses the problem of motion planning for robots that must satisfy temporal logic task specifications in environments where the semantic labels of regions are initially unknown. The robot must simultaneously learn the environment semantics and plan actions to fulfill its task specification, navigating the tension between exploration for information and exploitation of learned knowledge for task completion.

Code: github.com/taheri-az/Motion-Planning-Under-Temporal-Logic-Specifications-In-Semantically-Unknown-Environments

Download Paper