Autonomous Path Planning of AGV Obstacle Avoidance Based on Improved Q-learning

"Industry 4.0" and" Made in China 2025" have promoted the development of industrial automation equipment industry, and AGVs play a key role in warehousing intelligence and transportation efficiency. Research on AGV path planning is of great significance and application value for efficient, safe and high-quality path planning. Aiming at the problem of material transportation path planning in H company's static workshop, this paper proposes a path obstacle avoidance planning method based on improved Q learning algorithm. By optimizing Q learning parameters and introducing B-spline curve fitting technology to improve the algorithm, the AGV path is planned more effectively, and the operation efficiency and stability are improved. Through the analysis of simulation model verification results, the path length is 20.8995 meters before path smoothing and 21.2756 meters after path smoothing, with a small increase and little impact on work efficiency. Path smoothing reduces turning points and turning points, avoids path overlap and misalignment, and improves path efficiency. After the path smoothing improvement, the travel time was reduced from 15.8465 minutes to 10.1471 minutes, which was significantly reduced. Improve warehouse management level and transportation efficiency, and contribute to the development of industrial automation equipment industry.