Design and implementation of intelligent tracking car based on machine vision

Issue title: Special section: Intelligent data analysis and applications & smart vehicular technology, communications and applications

Guest editors: Valentina Emilia Balas and Lakhmi C. Jain

Article type: Research Article

Authors: Lisang, Liu^{a; b; c; d} | Dongwei, He^{a; c} | Jishi, Zheng^{a; b} | Ying, Ma^{a; c; *} | Jing, Huang^{a; c} | Junfeng, Fan^a | Xiaoyu, Wei^a | Jinxin, Yao^{a; c}

Affiliations: [a] Fujian University of Technology, Fuzhou, China | [b] FuJian Key Laboratory of A.E.D, Fujian University of Technology, Fuzhou, China | [c] National Demonstration Center for Experimental Electronic Information and Electrical Technology Education, Fuzhou, China | [d] School of Computer Science and Electronic Engineering, University of Essex, Wivenhoe Park, Colchester

Correspondence: [*] Corresponding author. Ma Ying, E-mail: *[email protected].

Abstract: The intelligent tracking car can realize self-perception, behavioral decision-making, automatic driving, etc. for environmental information, and has a wide application in our lives. In this paper, a new intelligent car is designed based on STM32F407ZGT6 MCU. The track information is collected by the OV7725 camera. The fast OTSU adaptive threshold algorithm is used to obtain the path guiding center line, which can realize image collection, image analysis and sensor data fusion, blocked path identification and intelligent judgment, automatic tracking function. The test results show that the algorithm is effective and feasible. When there are obstacles or partial loss on the track, it can intelligently identify the effective and obstacles and realize the independent decision making. It has good dynamic and robustness.

Keywords: Machine vision, intelligent car, tracking, obstacle avoidance

DOI: 10.3233/JIFS-179667

Journal: Journal of Intelligent & Fuzzy Systems, vol. 38, no. 5, pp. 5799-5810, 2020