Purpose-Path planning is an important part of UAV mission planning.The main purpose of this paper is to overcome the shortcomings of the standard particle swarm optimization(PSO)such as easy to fall into the local opt...Purpose-Path planning is an important part of UAV mission planning.The main purpose of this paper is to overcome the shortcomings of the standard particle swarm optimization(PSO)such as easy to fall into the local optimum,so that the improved PSO applied to the UAV path planning can enable the UAV to plan a better quality path.Design/methodology/approach-Firstly,the adaptation function is formulated by comprehensively considering the performance constraints of the flight target as well as the UAV itself.Secondly,the standard PSO is improved,and the improved particle swarm optimization with multi-strategy fusion(MFIPSO)is proposed.The method introduces class sigmoid inertia weight,adaptively adjusts the learning factors and at the same time incorporates K-means clustering ideas and introduces the Cauchy perturbation factor.Finally,MFIPSO is applied to UAV path planning.Findings-Simulation experiments are conducted in simple and complex scenarios,respectively,and the quality of the path is measured by the fitness value and straight line rate,and the experimental results show that MFIPSO enables the UAV to plan a path with better quality.Originality/value-Aiming at the standard PSO is prone to problems such as premature convergence,MFIPSO is proposed,which introduces class sigmoid inertia weight and adaptively adjusts the learning factor,balancing the global search ability and local convergence ability of the algorithm.The idea of K-means clustering algorithm is also incorporated to reduce the complexity of the algorithm while maintaining the diversity of particle swarm.In addition,the Cauchy perturbation is used to avoid the algorithm from falling into local optimum.Finally,the adaptability function is formulated by comprehensively considering the performance constraints of the flight target as well as the UAV itself,which improves the accuracy of the evaluation model.展开更多
基金supported by the Key Project of Science and Technology Innovation(2030)supported by the Ministry of Science and Technology of China(Grant No.2018AAA0101301)+2 种基金the Key Research Platforms and Projects of High School in Guangdong Province(No.2023ZDZX1028,2023ZDZX1050)Dongguan Social Development Science and Technology Project(No.20211800904722)Dongguan Science and Technology Special Commissioner Project(No.2021180050007).
文摘Purpose-Path planning is an important part of UAV mission planning.The main purpose of this paper is to overcome the shortcomings of the standard particle swarm optimization(PSO)such as easy to fall into the local optimum,so that the improved PSO applied to the UAV path planning can enable the UAV to plan a better quality path.Design/methodology/approach-Firstly,the adaptation function is formulated by comprehensively considering the performance constraints of the flight target as well as the UAV itself.Secondly,the standard PSO is improved,and the improved particle swarm optimization with multi-strategy fusion(MFIPSO)is proposed.The method introduces class sigmoid inertia weight,adaptively adjusts the learning factors and at the same time incorporates K-means clustering ideas and introduces the Cauchy perturbation factor.Finally,MFIPSO is applied to UAV path planning.Findings-Simulation experiments are conducted in simple and complex scenarios,respectively,and the quality of the path is measured by the fitness value and straight line rate,and the experimental results show that MFIPSO enables the UAV to plan a path with better quality.Originality/value-Aiming at the standard PSO is prone to problems such as premature convergence,MFIPSO is proposed,which introduces class sigmoid inertia weight and adaptively adjusts the learning factor,balancing the global search ability and local convergence ability of the algorithm.The idea of K-means clustering algorithm is also incorporated to reduce the complexity of the algorithm while maintaining the diversity of particle swarm.In addition,the Cauchy perturbation is used to avoid the algorithm from falling into local optimum.Finally,the adaptability function is formulated by comprehensively considering the performance constraints of the flight target as well as the UAV itself,which improves the accuracy of the evaluation model.
