Patel, Ruchir (University of Maryland, College Park), Rudnick-Cohen, Eliot (University of Maryland, College Park), Azarm, Shapour (University of Maryland, College Park), Herrmann, Jeffrey (University of Maryland)

Robust Multi-UAV Route Planning Considering UAV Failure

Scheduled for presentation during the Regular Session "Path Planning II" (WeB1), Wednesday, June 12, 2019, 15:20−15:40, Heritage B

2020 International Conference on Unmanned Aircraft Systems (ICUAS), June 11-14, 2019, Athens, Greece

This information is tentative and subject to change. Compiled on April 25, 2024

Abstract

This paper describes a robust multi-UAV route planning problem in which any one of the vehicles could fail during plan execution at any visited location. The UAVs must visit a set of fixed locations; if one UAV fails, the other vehicles must cover any unvisited locations. The objective is to optimize the worst-case cost. This paper formulates the problem with a min-sum objective (minimizing the total distance traveled by all vehicles) and a min-max objective (minimizing the maximum distance traveled by any vehicle). A Genetic Algorithm (GA) was used to find approximate robust optimal solutions on seven instances. The results show that the GA was able to find solutions that have better worst-case cost than the solutions generated by other approaches that were tested.