A parallel particle swarm optimization and enhanced sparrow search algorithm for unmanned aerial vehicle path...

Area of Science:

• Robotics and Automation
• Artificial Intelligence and Machine Learning
• Optimization Algorithms

Background:

• Unmanned Aerial Vehicle (UAV) path planning is critical for autonomous navigation.
• Traditional algorithms often yield suboptimal paths in complex environments.
• Enhancing path planning efficiency and effectiveness is an ongoing research challenge.

Purpose of the Study:

• To propose a novel hybrid algorithm, PESSA (Particle Swarm Optimization and Enhanced Sparrow Search Algorithm), for UAV path planning.
• To improve the global search capability and convergence speed of existing optimization algorithms.
• To validate the performance of PESSA in diverse 2D and 3D environments.

Main Methods:

• Developed PESSA by integrating Particle Swarm Optimization (PSO) with an Enhanced Sparrow Search Algorithm (ESSA).
• Enhanced ESSA with strengthened producer random jumps, scrounger learning from producer experience, and threat-adaptive search acceleration.
• Incorporated an elite reverse search strategy for optimal diversity and validated using 10 basic functions and 4 environmental scenarios.

Main Results:

• PESSA successfully found optimal values for 7 out of 10 test functions.
• Outperformed 12 other algorithms, consistently ranking first in average performance.
• Achieved average optimization results of 0.0165 and 0.0521 in 2D environments, and 0.6635 and 0.5349 in 3D environments.

Conclusions:

• The PESSA algorithm demonstrates superior performance in UAV path planning compared to existing methods.
• PESSA generates more feasible and effective flight routes across various complex environments.
• The enhanced ESSA component significantly contributes to improved global search and convergence.