A Lie group-based hybrid optimization framework for multi-objective UAV path planning using L-VGWO.

Area of Science:

• Robotics
• Artificial Intelligence
• Control Systems

Background:

• Three-dimensional UAV path planning faces challenges in precision, dynamic smoothness, and multi-objective optimization.
• Complex environments exacerbate these difficulties, requiring advanced algorithmic solutions.

Purpose of the Study:

• To propose a novel hybrid optimization framework, the Lie Group-based Griffon Vulture Grey Wolf Hybrid Optimizer (L-VGWO), for UAV path planning.
• To address high-precision pose alignment, dynamic path smoothness, and efficient multi-objective optimization.

Main Methods:

• Utilizes Lie group rotation-vector representation for path parameterization, preserving geometric continuity and avoiding singularities.
• Integrates Grey Wolf Optimizer (GWO) for local refinement and Griffon Vulture Optimizer (GVOA) for global exploration in L-VGWO's core solver.
• Employs a comprehensive objective function including obstacle avoidance, pose accuracy, and velocity smoothness constraints.

Main Results:

• The L-VGWO framework demonstrates improved balance between exploration and exploitation for multi-objective cost functions.
• Validated through multi-scenario simulations and statistical analyses, showing effectiveness in complex environments.
• Achieved high-precision pose alignment and dynamic path smoothness.

Conclusions:

• The L-VGWO framework offers an effective solution for challenging 3D UAV path planning problems.
• The hybrid optimization approach enhances performance in terms of precision, smoothness, and efficiency.
• Lie group parameterization contributes to robust rotational motion handling.