Read the full-text of this article for free

Title: An obstacle avoidance path selection for autonomous vehicles based on multi-dimensional data mining

Authors: Aiju Wang; Yao Yao; Zhanlei Shang

Addresses: College of Information Engineering, Zhengzhou University of Technology, Zhengzhou, 450044, China ' College of Information Engineering, Zhengzhou University of Technology, Zhengzhou, 450044, China ' Engineering Training Center, Zhengzhou University of Light Industry, Zhengzhou, 450001, China

Abstract: In order to overcome the problems of poor obstacle avoidance path selection, low success rate, and long time in traditional methods, a new obstacle avoidance path selection method for autonomous vehicles based on multi-dimensional data mining is proposed. The method employs the K-means algorithm to process multi-sensor data (including visual cameras, LightLaser Detection and Ranging (LiDAR), global positioning system (GPS), and traffic flow) for environmental data collection in autonomous vehicles. Based on the collected data and constraints, a target function for obstacle avoidance path selection of unmanned vehicles is constructed. The optimisation function is solved using the whale optimisation algorithm (WOA), and the optimal solution obtained is the obstacle avoidance path selection scheme for unmanned vehicles. Experimental results show that the proposed method for autonomous vehicle lane changing has a relatively large angle and short path, without collision problems. The maximum success rate of obstacle avoidance path selection is 98.56%, and the minimum time is 0.44 s.

Keywords: multi-dimensional data mining; autonomous vehicles; obstacle avoidance path selection; K-means algorithm; objective function; killer whale hunting algorithm.

DOI: 10.1504/IJVD.2025.147128

International Journal of Vehicle Design, 2025 Vol.97 No.5, pp.1 - 21

Received: 03 Dec 2024
Accepted: 02 May 2025
Published online: 10 Jul 2025 *