Multi-objective UAV Routing in Discretized and Continuous Spaces

Diclehan Tezcaner Öztürk

We consider the route planning problem for an Unmanned Air Vehicle (UAV) tasked with visiting multiple targets in a protected terrain. We form the routes considering three objectives; minimization of distance and radar detection threat, and maximization of the information collected from the targets visited. The UAV is free to move to any point within the terrain. This routing problem in the presence of multiple targets, multiple objectives, and continuous movement assumption is a complex problem. Between any two targets there is a continuum of efficient trajectory options, and finding a route requires determining both the visiting order to the targets and the trajectory to be used between consecutive target pairs. We develop a continuous terrain formulation allowing the trajectory to be selected from a continuum of trajectory options. We compare this model with its discretized version, in which the route is formed by selecting the trajectory to follow from a few of the trajectory options. In majority of the cases, the continuous terrain model yields solutions that outperform those of the discretized terrain model. For the continuous terrain model, we also develop an approximation algorithm that significantly reduces its computational requirements and ensures a desired level of precision. We show that the algorithm results in an approximate nondominated frontier that is sufficiently close to the true nondominated frontier.

Short Bio

Diclehan Tezcaner Öztürk is an Associate Professor in the Department of Industrial Engineering at Hacettepe University. She holds BSc, MSc, and PhD degrees from the Industrial Engineering Department at Middle East Technical University. She worked as a postdoctoral researcher at the University of Strathclyde. Her primary research interests include multiobjective decision making and logistics, with a particular focus on multiobjective combinatorial optimization problems.

Venue

Friday, October 31, 2025, 4.00 pm

IE Building, Halim Doğrusöz Auditorium (IE 03)