A Multiobjective Hybrid Optimization Algorithm for Path Planning of Coal Mine Patrol Robot

Area of Science:

• Robotics
• Artificial Intelligence
• Mine Safety Engineering

Background:

• Coal mine patrol robots face challenges with long, inefficient paths due to poor visibility and road conditions.
• Existing path planning algorithms struggle to navigate complex underground environments effectively.
• The need for accurate and efficient navigation is critical for robot safety and operational success.

Purpose of the Study:

• To develop an improved hybrid path planning algorithm for coal mine patrol robots.
• To enhance path planning accuracy and obstacle avoidance capabilities in challenging underground conditions.
• To optimize route length, reduce travel time, and improve navigation efficiency.

Main Methods:

• Integration of an improved Artificial Fish Swarm Algorithm (AFSA) and Dynamic Window Algorithm (DWA) for global path planning.
• Introduction of an improved Genetic Algorithm (GA) to enhance path planning accuracy.
• Development of an adaptive trajectory evaluation function within the improved DWA for local obstacle avoidance.
• Establishment of a simulation platform using Robot Operating System (ROS).

Main Results:

• The proposed algorithm demonstrated feasibility through simulations.
• Simulations showed a reduction in path length by 0.12m and time by 3.14s.
• The algorithm effectively removed turning points and redundant path segments, leading to smoother trajectories.
• Improved obstacle avoidance capabilities were observed.

Conclusions:

• The improved hybrid path planning algorithm is effective and superior for coal mine patrol robots.
• The algorithm enhances navigation efficiency and safety in complex underground environments.
• The developed simulation platform validates the practical applicability of the proposed method.