According to back-to-back traversal of plant protection UAV, a kind of working area decomposing method based on grid method was used to extract the UAV working path. A route planning algorithm with the minimum return number was proposed in order to reduce the ineffective energy consumption in non-operate situation. The spraying amount and return points of the sorties were reasonably allocated to improve operational efficiency. The simulation results show that using this algorithm in an area of 15600m2, compared to the return route in the liquid exhaustion situation, energy consumption of the UAV was reduced by 5.99%; while in an area of 42000m2, energy consumption of the UAV was even reduced by 12.89%. When the operation area was larger, the energysaving effect was more obvious, which proved the feasibility of route planning algorithm. Field tests in an area of 2500m2 proved that there were deviations between theory and practical routes. The possible reasons were GPS positioning error, wind and center of gravity in liquid. So the GPS positioning accuracy needs to be improved, and the flight control system needs to be further improved. This flight route planning algorithm was proposed for automatic UAVs in unmanned operation, thus the flight routes were set before operation, and the energy consumption of UAVs and flight sorties can be estimated in advance. Nowadays, the unmanned operation becomes a new trend, and this flight route planning algorithm can be widely used in precision agriculture.