Unmanned Aerial Vehicles (UAVs) represent an important asset in the increasing automation required in Smart Cities. However, their short flight autonomy and limited computational resources raise a set of issues to unravel for their full exploitation. In this paper, some of these issues are addressed by leveraging the deployment of Training and Recharging Areas (TRA) in the Smart City. While moving among different Mission Areas (MA), the UAVs pass through dedicated TRA, where Energy and Data Dispensers (EDD) are scattered to allow UAV battery recharge and software update. The tridimensional trajectory planning problem to let the UAVs efficiently and swiftly move through the EDDs, towards the exit of the TRA, is investigated. A mathematical formulation of the problem is presented, and an online approach is designed to achieve a feasible implementation in a realistic scenario.