Graduation Semester and Year
Fall 2025
Language
English
Document Type
Dissertation
Degree Name
Doctor of Philosophy in Computer Science
Department
Computer Science and Engineering
First Advisor
Manfred Huber
Second Advisor
Vassilis Athitsos
Third Advisor
Farhad A. Kamangar
Fourth Advisor
Habeeb Olufowobi
Abstract
Unmanned Aerial Systems (UAS) have become increasingly popular as versatile platforms for tasks such as surveillance, inspection, delivery, and maintenance. In many applications, UAS operate in environments frequented by people or containing sensitive infrastructure, which introduces physical risks in case of vehicle failure, as well as psychological and privacy concerns that may limit their acceptability. Ensuring safe and efficient operation thus requires that UAS consider these risks when planning navigation strategies. While prior information, such as city maps and building layouts, can partially inform risk assessment, such data is often incomplete, necessitating real-time augmentation of risk maps using sensor information. To address this, I propose an approach that leverages learned risk identification from aerial imagery to dynamically fuse new information with prior data, producing an updated risk map that enables effective re-planning of navigation strategies. Achieving this in dynamic environments often requires coordination between multiple agents, which is analogous to human collaboration: we infer the intentions of our peers from environmental changes and select actions that efficiently achieve shared goals. For artificial intelligent agents, predicting intentions and coordinating in partially observable environments is challenging, especially in distributed systems where direct communication may be limited or unavailable. In this work, I present an intention-aware algorithm that allows multiple agents to independently make decisions based on observations and predicted behaviors of others, enabling safe, efficient, and collaborative navigation in complex, uncertain environments.
Keywords
Risk Map Fusion, Dynamic Path Re-planning, Dynamic Multi RRT, Intention aware multi agent reinforcement learning, Intention prediction in partially observable environment, Deep learning in dec-pomdp, Unmanned vehicle safe navigation, risk map fusion for unmanned vehicle
Disciplines
Artificial Intelligence and Robotics | Robotics
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Recommended Citation
Trivedi, Bhaskar Chandra, "Environment and Intention Awareness for Navigation and Collaboration" (2025). Computer Science and Engineering Dissertations. 425.
https://mavmatrix.uta.edu/cse_dissertations/425