An improved chaos sparrow search algorithm for UAV path planning.

Area of Science:

• Artificial Intelligence
• Robotics
• Optimization Algorithms

Background:

• Path planning for Unmanned Aerial Vehicles (UAVs) in 3D complex environments faces challenges with slow convergence and local optima.
• Existing algorithms like Sparrow Search Algorithm (SSA) and Particle Swarm Optimization (PSO) may not sufficiently address these issues.

Purpose of the Study:

• To propose an improved chaos sparrow search algorithm (CSSA) for efficient and accurate UAV 3D path planning.
• To enhance the algorithm's convergence speed and ability to escape local optima in complex environments.

Main Methods:

• Population initialization using piecewise chaotic mapping to improve initial solution quality.
• Introduction of a nonlinear dynamic weighting factor for producer updates to balance exploration and exploitation.
• Integration of an enhanced sine cosine algorithm for scrounger updates to broaden search space.
• Application of dynamic boundary lens imaging reverse learning to prevent local optima entrapment.

Main Results:

• The proposed algorithm demonstrated superior performance compared to CSSA, SSA, and PSO in UAV 3D path planning.
• Significant time improvements were observed in complex environments: 22.4% over CSSA, 28.8% over SSA, and 46.8% over PSO.
• The algorithm exhibited high convergence accuracy, confirming its effectiveness.

Conclusions:

• The improved CSSA effectively addresses the limitations of existing algorithms in UAV 3D path planning.
• The enhancements lead to faster convergence, better exploration, and robust avoidance of local optima.
• The proposed algorithm is a valuable and superior solution for complex UAV navigation tasks.