UAV Path Planning for Data Ferrying with Communication Constraints

Jose Joseph, Mohammadreza Radmanesh, Mohammad Nazmus Sadat, Rui Dai, Manish Kumar

Abstract: Unmanned aerial vehicles (UAVs) have been used in many surveillance and remote sensing applications due to their capabilities of flying over large areas with sensing devices. The sensed data such as weather information or surveillance footage is usually very time sensitive, and the mission will be successful only if the data could be captured and delivered to processing centers on time. Using the data ferrying mechanism, UAVs could collect sensed data from remote areas and ferry the data to appropriate locations for offloading before the on-board storage spaces are overloaded. The data collection and offloading strategies of UAVs should be carefully determined to ensure the timely delivery of information while saving the travel costs. This paper investigates a UAV path planning problem considering realistic data ferrying situations involving data offloading time windows and communication constraints for periodic remote sensing applications. A Mixed Integer Linear Programming (MILP) problem is formulated that finds paths for UAVs with the minimal cost to meet application deadline requirements under time window and communication constraints on data collection and offloading. A genetic algorithm (GA) based approach is designed to solve this problem. Evaluation results on different input parameters show that the proposed approach could produce near-optimal solutions with much faster execution time than the brute-force approach. The scalability of the proposed approach is validated by a major test case with 120 waypoints, in which stabilized near-optimal solutions could be found within 60,000 iterations.