Global Dynamic Path Planning of AGV Based on Fusion of Improved A* Algorithm and Dynamic Window Method

Area of Science:

• Robotics
• Artificial Intelligence
• Computer Science

Background:

• Automated Guided Vehicles (AGVs) require efficient global path planning and real-time dynamic obstacle avoidance.
• Traditional algorithms face challenges in complex, dynamic environments with unpredictable obstacles.

Purpose of the Study:

• To develop a fusion algorithm combining an improved A* algorithm and the dynamic window method for AGV global optimal path planning and dynamic obstacle avoidance.
• To enhance path planning efficiency, reduce search scope, and improve obstacle avoidance capabilities.

Main Methods:

• Dynamically weighting the heuristic function of the A* algorithm to reduce search scope.
• Implementing a path-optimization method to eliminate redundant nodes and turning points.
• Integrating the improved A* algorithm with the dynamic window method for local dynamic obstacle avoidance within the global path.

Main Results:

• The improved A* algorithm demonstrated a 26.3% reduction in planning time and a 57.9% smaller search scope compared to the traditional A* algorithm.
• Path length was reduced by 7.2%, with significant reductions in path nodes (85.7%) and turning points (71.4%).
• The fusion algorithm successfully evaded moving and unknown static obstacles in real-time within diverse map environments.

Conclusions:

• The proposed fusion algorithm effectively addresses AGV global path planning and dynamic obstacle avoidance demands.
• The enhanced A* algorithm significantly improves planning efficiency and path quality.
• The integrated approach ensures real-time navigation and obstacle evasion in complex, dynamic environments.