The field of Obstacle Avoidance and Navigation (OAN) algorithms is constantly being researched, with innovations having far‐reaching uses. These methods can be generally split into two variants – offline methods and online methods. Offline methods tend to require a prior known map to function but provide goal‐reaching guarantees, while online methods forego this need but may result in getting stuck on local minima. Many OAN schemes that currently exist are usually a compromise of either computational efficiency or algorithm robustness. However, when integrating Obstacle Avoidance Navigation (OAN) algorithms into commercially used robots, like multirotor unmanned aerial vehicles (UAVs), the algorithms need to be efficient and robust while also being light enough computationally to operate directly on the UAV. This section introduces one such algorithm, the Closest Obstacle Avoidance and A* (COAA*) algorithm, which addresses the limitations associated with both offline and online algorithms while ensuring it reaches the most optimal solution. This algorithm has been tested on the Heavy Lift Experimental (HLX) UAV at Taylor's University, with the simulation results closely mirroring those obtained in real‐world scenarios, underlining its potential for broad application.
Enhancing UAV Navigation in Partially Observable 2D Environments
An Optimized Obstacle Avoidance Approach
24.06.2025
29 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Decision-Making for Autonomous Systems in Partially Observable Environments
BASE | 2013
|Motion planning for autonomous vehicles in partially observable environments
TIBKAT | 2023
|Motion Planning for Autonomous Vehicles in Partially Observable Environments
TIBKAT | 2023
|Motion planning for autonomous vehicles in partially observable environments
TIBKAT | 2022
|Motion Planning for Autonomous Vehicles in Partially Observable Environments
GWLB - Gottfried Wilhelm Leibniz Bibliothek | 2023
|