Abstract

Autonomous mobile robots are increasingly expected to perform complex missions in unstructured environments. Traditional path planning approaches handle simple point-to-point navigation, but struggle with complex tasks that involve temporal and logical orderings of objectives. Linear Temporal Logic (LTL) provides a method for complex missions (e.g., sequential visits to multiple targets or surveillance tasks) in a strict way. This thesis presents an integrated planning framework that enables a robot to satisfy LTL-based task specifications in an unknown environment by combining a Temporal Logic RRT* (TL-RRT*) planner with semantic mapping. The robot builds a semantic map of its environment online using simultaneous localization and mapping (SLAM) with object-level segmentation, so that high-level task objectives grounded in semantic labels can be understood and achieved. A replanning mechanism is implemented to continually update the robot plan as new obstacles and landmarks are discovered, ensuring the LTL mission can be fulfilled despite the initially unknown environment. The system architecture is implemented in ROS2, with modules for perception, mapping, planning, and control. We evaluated the proposed system in realistic simulation scenarios with a mobile robot navigating an indoor environment containing various objects. The results demonstrate that the robot can successfully perform temporal logic missions, such as visiting multiple goal regions in order or avoiding forbidden areas, by leveraging the semantic map and on-line replanning. We report the success rates for a range of task specifications, showing the effectiveness of biasing the planner with semantic knowledge. Finally, we discuss the advantages of the approach, its current limitations (such as partial observability and computational complexity), and potential directions for future research in formal task planning for autonomous robots.

Committee Chair

Yiannis Kantaros

Committee Members

Fanwei Kong Hussein Sibai

Degree

Master of Science (MS)

Author's Department

Mechanical Engineering & Materials Science

Author's School

McKelvey School of Engineering

Document Type

Thesis

Date of Award

Spring 5-2026

Language

English (en)

Download

Share

COinS