Research On UAV Path Planning Based On Improved Differential Evolution Algorithm

UAV(Unmanned Aerial Vehicle,UAV)has been widely used in military and civilian fields due to its safety,high efficiency,etc.The efficient of the mission execution mainly depends on the reliable and safe of the UAV flight path.That is difficult to design an efficient planning algorithm for solving the flight path of UAV in a mission scenario,especially in a highly complex flight environment.we have studied systematically on the method of the UAV flight planning in complex scenarios,including constructing flight scenario model,constructing flight constraint and cost functions,and designing optimization algorithms,etc.The main research contents and results are described as follows:A UAV trajectory planning model has been established.Based on the grid method,the virtual terrain of the three-dimensional environment for UAV flight is established with the mathematical functions.In the virtual terrain,the four scenarios are constructed with different complexity levels by changing the number and location of the threats,such as radar,artillery,and no-fly zone.Meanwhile,the UAV flight height and length are considered as the cost functions.The constraint model mainly contains UAV climb angle,turning radius,radar,artillery and no-fly zone.An improved differential evolution algorithm has been proposed to plan the UAV flight path.Based on the standard differential evolution algorithm,multi-population strategy,interactive mutation strategy and teaching-learning-based method for the self-adaptive strategy have been introduced to improve the algorithm robustness and convergence.The constraints are transformed into a single-objective optimization problem by the linear weighted sum method for the path planning process.It is resulted that the proposed algorithm has strong robustness and convergence in solving the three-dimensional trajectory planning.This algorithm is more superiority over other comparison algorithms as the complexity of the scene increasing.The feasible rule mechanism is adopted to deal with the constraints in the UAV path planning.In this work,the feasible rules is used to deal with the UAV physical constraints and threat constraints,and the objective function is normalized.It is resulted that the improved differential evolution algorithm based on the feasible rules has stronger search ability,convergence ability and stability for feasible solutions in the multi-constrained UAV trajectory planning problems.