Planning of takeoff/landing site location, dispatch route, and spraying route for a pesticide application helicopter

Abstract

Relying on the pilot's flight experience to plan a temporary helicopter takeoff/landing site location, dispatch routes, and spraying routes often results in very long dispatch routes, imprecise forest area coverage, and a high extra coverage rate. An intelligent fusion algorithm was proposed to address these problems. A vector modeling method (VMM) was proposed to determine the optimum spraying route with the minimum number of turnarounds and the lower extra coverage rate. An improved genetic algorithm (IGA) was used to plan the dispatch route. Subsequently, the improved SSA (ISSA) was utilized to plan the takeoff/landing site location. An artificial neural network (ANN)-based classifier was constructed to constrain the search domain of the ISSA to avoid ponds, lakes, and other areas unsuitable for takeoff/landing. The proposed intelligent fusion algorithm is called VMM-ISSA-IGA-ANN. The planning of the takeoff/landing site locations, dispatch routes, and spraying routes for helicopter plant protection operations was carried out in four forested environments in Zhengzhou, Kaifeng, Mingguang, and Jurong, China. The planning scheme was compared with that of an aviation company. The results showed that the takeoff/landing site location derived from the VMM-ISSA-IGA-ANN was more reasonable than that obtained by the aviation company. The dispatch route was shortened by 6.51%, 23.7%, 42.07%, and 49.80%, the spraying route was shortened by 1.94%, 0.22%, 0.05%, and 0.91%, the extra coverage rate was reduced by 29.15%, 24.92%, 14.46%, and 26.04%, and the number of turnarounds was reduced by 7, 2, 21, and 22 for the four locations, respectively. This study provides a theoretical basis for route planning of helicopter aerial spraying operations and a technical reference for smart forestry applications.