In recent years, unmanned aerial vehicle (UAV) technology has made remarkable progress in many fields, especially in the path planning of fixed-wing UAVs. In the face of complex task scenarios, fixed UAVs need to consider various scenarios, generate conforms to aircraft performance, and make autonomous decision-making to evade and standby flight routes to meet the needs of load adaptation and performance guarantee. Especially for some specific load unmanned aerial vehicles (UAVs), the flight path planning is in advance before launch, so the ground stage of route planning is the key. This paper puts forward A kind based on A* heuristic search algorithm of the optimum path planning method, through the Alpha - Beta pruning to accelerate the search process, using Douglas-Peucker curve smoking loose to reduce cost, time and space combined with spline interpolation and polynomial fitting rounds of optimization, Finally, the critical routes with the ability to evade, wait and attack are generated. The experimental results show that the algorithm achieves 4.5% optimization in the path length, and the smoothness of the curve is improved by 36.71%. It has the planning ability to hover and wait in the mission area, effectively improving the path generation’s smoothness, effectiveness, and refinement. It also reduces the route length, showing good anti-data noise and long path planning ability.