UAV Mesh Network Trajectory Planning for Age Optimal Data Collection in Infrastructureless Areas

Collection of environmental data from emergency sites where there is no or minimal cellular infrastructure exists is very critical for efficient response management. Unmanned aerial vehicles (UAV) can provide a tremendous support during that process thanks to their flexibility, agility and lower cost. A mesh network formed among the UAVs can facilitate the data collection process while also keeping the communication among them. However, as the age of the collected information (i.e., from the moment it is generated at the ground sensor node to the moment it is delivered to the emergency response center) defines the success of response tasks, the trajectory of the UAV mesh network should be determined carefully considering the timely delivery of the critical data. In this paper, we study the path planning problem for a UAV mesh network for AoI optimal data collection from the ground IoT devices in such emergency sites with minimal or no infrastructure. We explore different settings that could happen in such scenarios and develop an Integer Linear Programming (ILP) based model for each to optimize the UAV trajectories with the main goal of minimizing the maximum AoI from the collected data. In order to avoid the high complexity of ILP solutions, we also propose relaxed models. Through simulations, we compare the results in different scenarios in terms of the maximum AoI and UAV path lengths and discuss potential drawbacks in each.