Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List

Full text release has been delayed at the author's request until August 03, 2025

ETD Abstract Container

Abstract Header

Noise Aware Hybrid Fuel UAV Path Planning and Power Management

Scott, Drew
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1721229697879083

Abstract Details

2024, PhD, University of Cincinnati, Engineering and Applied Science: Mechanical Engineering.
The path planning and power management of hybrid fuel UAVs under presence of noise-restrictions is studied here. This problem is motivated by two scenarios: i) widespread use of UAVs in congested, urban environment; and ii) Noise-sensitive surveillance missions. In either case, it is envisioned that noise-restrictions are in place in subsets of the environment, such that ground-level noise produced by the UAV at hand must be under a certain intensity. In the case of urban usage, we consider it likely that such restrictions are eventually put in place near residential and business areas. In the case of a hybrid-fuel UAV, where energy sources include a battery-pack and combustion engine, the noise produced by the engine is intense relative to the propeller noise. In this scenario, the path planning and power planning is a coupled problem: given a path, certain power plans are infeasible, and given an energy plan certain paths are infeasible. Thus, the path of the UAV must be found in tandem with the power plan. This results in a novel problem, which we study here. The single-agent problem is studied first within a discrete framework, as is standard for vehicle motion planning. An environment is discretized into a graph, such that nodes represent locations in the configuration space and edges between the nodes are flight legs the UAV travels along between nodes. Edges are parameterized by cost and energy values. The objective is to find a feasible sequence of nodes of lowest cost without violating the power and noise constraints. We develop a fast, exact algorithm to solve this planning problem quickly on graphs of tens of thousands of nodes. The problem is approached in an optimal control framework, with only an initial approach presented in this dissertation. Battery modeling in the context of this problem is also studied briefly. The final piece of work is returning to the discrete problem in the context of multi-agent path finding (MAPF). The standard MAPF seeks to find a set of paths for each agent in the system such that no conflicts between any two paths exist. We extend this problem to consider hybrid-fuel vehicles and noise-restrictions as in the single agent problem, which we dub the Noise-Restricted Hybrid-Fuel MAPF (NRHF-MAPF). A recent approach to MAPF is Conflict Based Search (CBS) which makes use of a constraint tree to find the optimal solution. A heuristic variant is Enhanced CBS (ECBS). We implement CBS and ECBS to solve our NRHF-MAPF. The latter requires a heuristic subroutine to quickly find feasible solutions. We present a heuristic variant of the labeling algorithm discussed above for the single-agent subproblem which arises in ECBS and CBS, which we refer to as the FOCAL labeling algorithm. It is shown that despite the NP-Hard subproblem, the labeling algorithm utilized within CBS and ECBS allows large NRHF-MAPF instances to be solved. This extension of the original labeling algorithm for the NRHFSPP demonstrates the robustness and utility of the label-correcting paradigm.
Manish Kumar, Ph.D. (Committee Chair)
David Casbeer, Ph.D M.A B.A. (Committee Member)
Kenny Chour, Ph.D M.A B.A. (Committee Member)
Michael Alexander-Ramos, Ph.D. (Committee Member)
David Thompson, Ph.D. (Committee Member)
Satyanarayana Gupta Manyam, Ph.D M.A B.A. (Committee Member)
146 p.
Operations Research
UAV; path planning; noise aware; optimization; hybrid-fuel; MAPF

Recommended Citations

Citations

  • Scott, D. (2024). Noise Aware Hybrid Fuel UAV Path Planning and Power Management [Doctoral dissertation, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1721229697879083

    APA Style (7th edition)

  • Scott, Drew. Noise Aware Hybrid Fuel UAV Path Planning and Power Management. 2024. University of Cincinnati, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1721229697879083.

    MLA Style (8th edition)

  • Scott, Drew. "Noise Aware Hybrid Fuel UAV Path Planning and Power Management." Doctoral dissertation, University of Cincinnati, 2024. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1721229697879083

    Chicago Manual of Style (17th edition)

ucin1721229697879083
© 2024, all rights reserved.
This open access ETD is published by University of Cincinnati and OhioLINK.