Research on Mobile Robot Path Planning Based on MSIAR-GWO Algorithm.

Area of Science:

• Robotics
• Artificial Intelligence
• Optimization Algorithms

Background:

• Path planning is crucial for mobile robot navigation efficiency and safety.
• Traditional Gray Wolf Optimization (GWO) faces challenges like slow convergence and local optima.
• Balancing exploration and exploitation remains difficult in practical GWO applications.

Purpose of the Study:

• To propose a Multi-Strategy Improved Gray Wolf Optimization (MSIAR-GWO) algorithm for enhanced path planning.
• To address the limitations of traditional GWO, including slow convergence and local optima.
• To improve the efficiency and safety of mobile robot autonomous navigation.

Main Methods:

• Introduced a nonlinear convergence factor with reinforcement learning for intelligent parameter configuration.
• Implemented an adaptive position-update strategy with detour foraging and dynamic weights.
• Incorporated an artificial rabbit optimization algorithm bypass foraging strategy with Brownian motion and Levy flight.
• Utilized an elimination and relocation strategy based on stochastic center-of-gravity dynamic reverse learning for inferior individuals.

Main Results:

• MSIAR-GWO demonstrated excellent stability, higher solution accuracy, and faster convergence on benchmark test functions.
• The algorithm effectively planned shorter and smoother paths in complex raster map environments.
• Compared to traditional algorithms, MSIAR-GWO showed superior optimization-seeking ability and robustness.

Conclusions:

• The proposed MSIAR-GWO algorithm significantly improves upon traditional GWO for path planning.
• MSIAR-GWO offers a robust and efficient solution for mobile robot autonomous navigation.
• The integration of multiple strategies enhances convergence speed, accuracy, and global search capabilities.