A spherical vector-based adaptive evolutionary particle swarm optimization for UAV path planning under threat...

Area of Science:

• Robotics and Control Systems
• Artificial Intelligence and Optimization Algorithms
• Aerospace Engineering

Background:

• Unmanned Aerial Vehicle (UAV) path planning is critical for complex missions.
• Existing Particle Swarm Optimization (PSO) methods often fail due to inadequate consideration of UAV dynamics, leading to infeasible paths and suboptimal solutions.
• Efficient and safe path planning is essential for the expanding scale of UAV applications.

Purpose of the Study:

• To develop an advanced optimization algorithm that effectively addresses the constraints and complexities of UAV path planning.
• To improve the efficiency, safety, and accuracy of UAV path planning compared to existing methods.
• To enhance the convergence stability and scalability of path planning algorithms in dynamic and challenging environments.

Main Methods:

• Proposed a spherical vector-based adaptive evolutionary particle swarm optimization (SAEPSO) algorithm.
• Integrated UAV dynamic constraints directly into the spherical vector framework.
• Incorporated improved tent map and reverse learning for initial solution diversity and distribution.
• Implemented dynamic nonlinear and adaptive factors to balance exploration and exploitation.
• Introduced an adaptive acceleration strategy for underperforming particles and an evolutionary programming strategy.

Main Results:

• The SAEPSO algorithm demonstrated superior performance in initial solution effectiveness and final solution accuracy.
• Achieved enhanced convergence stability and scalability across various benchmark scenarios with different threat levels.
• Outperformed existing algorithms in addressing UAV dynamic constraints and complex environmental challenges.
• Generated more feasible and optimal paths compared to traditional PSO methods.

Conclusions:

• The proposed SAEPSO algorithm offers a significant advancement in UAV path planning optimization.
• SAEPSO effectively handles dynamic constraints and complex environments, leading to more reliable and efficient UAV missions.
• The algorithm's enhanced diversity, exploration/exploitation balance, and adaptive strategies contribute to its robust performance and scalability.