Fair Formation Control of Multiple Agents Using Ant Colony Optimization
Yoshie Suzuki, Stephen Raharja, Toshiharu Sugawara · 2022 Joint 12th International Conference on Soft Computing and Intelligent Systems and 23rd International Symposium on Advanced Intelligent Systems (SCIS&ISIS) · 2022
This paper proposes a method to automatically generate consecutive formation transition paths of multiple agents, such as in unmanned aerial vehicle (UAV) shows, considering the reduced total travel distance as well as fair travel distances of individual agents. A UAV show is a type of entertainment where various artistic formation patterns are presented in the air by UAV agents, such as drones. Although several studies discussed the formation transition for UAV shows by considering the minimization of the total travel distance, few of them considered the fairness in moving distances among agents and transitions for long distances, which result in battery runout and shorter performance times Moreover, as a UAV show is performed by several agents, their battery capacities are usually not large; however, the performance time is affected by battery consumption. Our proposed method, which is based on ant colony optimization (ACO), enables agents to present longer performances in 3D space by considering the fairness in travel distance. Experiments showed that agents could perform approximately 20% more formation patterns without collisions with our proposed method based on ACO than those with the conventional method that is also based on ACO.