Date of Award

8-2026

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Automotive Engineering

Committee Chair/Advisor

Dr. Robert Prucka (Co-Chair)

Committee Member

Dr. Qilun Zhu (Co-Chair)

Committee Member

Dr. Jiangfeng Zhang

Committee Member

Dr. Benjamin Lawler

Committee Member

Dr. Zoran Filipi

Abstract

Autonomous off-road vehicles with electrified powertrains operate in unstructured terrain where grade, curvature, traction variability, and obstacle-induced detours strongly influence traction demand, battery loading, auxiliary thermal power, and component temperature evolution. Existing architectures typically decouple trajectory generation from energy and thermal management, or optimize powertrain states along a predetermined route, which can lead to inefficiency, accelerated degradation, and feasibility loss under demanding operating conditions. This dissertation develops a hierarchical predictive framework for integrated energy, thermal, and trajectory-aware planning for autonomous off-road vehicles. First, control-oriented electro-thermal models and a predictive battery thermal management controller are developed to exploit future load information, reducing cooling-system energy use by 55% relative to a no-preview optimal controller while establishing preview-horizon and battery-sizing design relations. Second, a model predictive Integrated Energy and Thermal Planner coordinates engine-generator operation and thermal actuation in a series hybrid tracked vehicle using a computationally efficient priority-speed formulation, reducing battery degradation by up to 29% and improving fuel efficiency by at least 10% relative to separated planning in demanding scenarios. Third, a curriculum-trained reinforcement-learning and model-predictive path-integral framework improves the computational tractability and robustness of off-road trajectory planning, achieving up to 70% higher success rates and up to 90% lower sample requirements than MPPI alone. Finally, the Integrated Propulsion, Thermal, and Route Planner unifies route execution, motion, propulsion, braking, and thermal actuation at the mission level and introduces a threshold-guided characterization of when trajectory-first planning becomes structurally suboptimal due to electro-thermal constraint tightening. In this tightening-prone regime, the proposed integrated planner improves realized mission performance by about 3-4% and improves the combined powertrain performance by about 30-35% relative to trajectory-first architectures, confirming the predictive value of the proposed threshold-based characterization. In shorter, lower-loading missions where constraint tightening does not activate, the performance gap narrows, consistent with the proposed characterization. Together, these contributions show that energy efficiency and thermal feasibility in autonomous off-road vehicles must be addressed jointly across subsystem, supervisory, and global planning layers.

Author ORCID Identifier

0009-0003-5927-1278

Download

Available for download on Tuesday, August 31, 2027

Share

COinS