Development of a self-driving car prototype for educational and research purposes. Issue 1 (May 2021)

• Record Type: Journal Article
• Authors: Fan, X; Zhang, Z; Wang, Y
• Is Part Of: Journal of physics. Volume 1905:Issue 1(2021)
• Publication Date: 2021-05
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1905/1/012013
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: Self-driving car (SDC) is one of the most challenging topics in artificial intelligence. However, current off-the-shelf SDC platforms are expensive and difficult to modify, which prevents more researchers from working towards the improvement of self-driving technologies. This paper presents the development of a SDC prototype with low-cost sensors and an open-source software platform for educational and academic research purposes. The goal of such a SDC prototype is to serve as a testbench, which allows students and researchers to easily test new algorithms of SDC. Localization is one of the most basic and important problems in SDC. To achieve high performance using low-cost sensors, a localization method based on Lidar-INS fusion was investigated under the framework of Kalman filtering. Autonomous driving experiments were conducted on roads of a university campus environment and the results are used to show the performance.