Title: A method of environment perception for automatic driving tunnel based on multi-source information fusion
Authors: Jie Luo
Addresses: Intelligent Manufacturing and Automobile School, Chongqing College of Electronic Engineering, Chongqing 401331, China
Abstract: In order to improve the collection accuracy of multi-source information, reduce the false alarm rate of tunnel driving risk and improve the accuracy of environmental perception, a new tunnel environment awareness method based on multi-source information fusion is proposed in this paper. Firstly, R-Fans-16 line lidar and GP-KD6Q01FC monocular camera are selected as the information acquisition equipment. Secondly, according to the multi-source information fusion principle, the Markov distance between the observation value and the actual value is calculated to complete the multi-source data fusion. Finally, the fused information is input into the Adaboost classifier to complete the accurate classification perception of the autonomous driving tunnel environment. The experimental results show that, compared with the traditional environment awareness methods, the proposed method can accurately collect multi-source information, and can achieve high-precision perception of tunnel environment, with the highest perception accuracy of 97.7%.
Keywords: multi-source information fusion; automatic driving; line lidar; tunnel environment perception; monocular camera.
DOI: 10.1504/IJCISTUDIES.2023.132499
International Journal of Computational Intelligence Studies, 2023 Vol.12 No.1/2, pp.142 - 154
Received: 25 Nov 2022
Accepted: 15 Feb 2023
Published online: 24 Jul 2023 *