Path Planning of an Unmanned Aerial Vehicle Based on a Multi-Strategy Improved Pelican Optimization Algorithm

Area of Science:

• Robotics and Automation
• Artificial Intelligence
• Aerospace Engineering

Background:

• Effective Unmanned Aerial Vehicle (UAV) path planning is crucial for optimizing tasks in complex urban environments.
• Existing algorithms often struggle with multi-objective optimization, balancing path length, turning angles, and collision avoidance.

Purpose of the Study:

• To develop an advanced UAV path planning algorithm that enhances efficiency and safety.
• To address the multi-constraint optimization challenge in UAV navigation.

Main Methods:

• A multi-strategy improved Path Optimization Algorithm (IPOA) was developed.
• IPOA integrates chaotic mapping, refracted reverse learning, nonlinear inertia weights, Levy flight, and adaptive t-distribution variation.
• The algorithm transforms path planning into a multi-constraint optimization problem considering path length, turning angle, and collision avoidance.

Main Results:

• IPOA demonstrated superior performance against other algorithms in 69.4% of CEC2022 test functions.
• In real-world simulations, IPOA improved path length by 8.44%, turning angle by 5.82%, obstacle avoidance by 4.07%, and flight time by 9.36% compared to the original POA.
• Compared to MPOA, IPOA showed improvements of 4.09% in path length, 0.76% in turning angle, 1.85% in obstacle avoidance, and 4.21% in flight time.

Conclusions:

• The proposed IPOA algorithm significantly enhances UAV path planning efficiency and accuracy.
• IPOA offers a robust solution for complex urban navigation tasks, outperforming existing methods.