We study route scheduling of a UAV for data collec-tion from remote sensor nodes (SNs) with battery recharging. Thefreshness of the collected information is captured by the metric ofage of information (AoI). The objective is to minimize the averageAoI cost of all SNs over a scheduling time horizon. We prove thatthe problem is NP-hard via a reduction from the Hamiltonianpath. Next, we prove tractability of the problem for a symmetricscenario. For problem solving, we develop an algorithm based ongraph labeling. Finally, we show the effectiveness of our algorithmin comparison to greedy scheduling.