Title: Visual automatic obstacle avoidance technology research in unmanned vehicles

Authors: Bo Liu; Liguang Li; Piqiang Tan; Rui Jia; Qing Liu

Addresses: School of Automotive Studies, Tongji University, Shanghai, 201804, China; Shanghai Kakuka New Energy Technology Co., Ltd., Shanghai, 200082, China ' School of Automotive Studies, Tongji University, Shanghai, 201804, China ' School of Automotive Studies, Tongji University, Shanghai, 201804, China ' Shanghai Kakuka New Energy Technology Co., Ltd., Shanghai, 200082, China ' Shanghai Kakuka New Energy Technology Co., Ltd., Shanghai, 200082, China

Abstract: In view of the problems of low obstacle avoidance and low efficiency in traditional unmanned vehicle in automatic obstacle avoidance, multi-feature fusion automatic obstacle avoidance method in unmanned vehicle is proposed. Optimise unmanned vehicle obstacle avoidance objective function, measure target obstacle distance, and calculate the braking distance. Based on this, unmanned vehicle dynamics model is established and the obstacle is located and determined. Multi-feature fusion design of unmanned vehicle visual automatic obstruction steps is made and finally the process of obstacle avoidance is analysed. The experimental results show that the use of improved visual automatic obstacle avoidance technology has certain advantages in target obstacle positioning and obstacle avoidance accuracy, which are superior to those of traditional obstacle avoidance technology.

Keywords: vehicle; unmanned; visual; automatic; obstacle avoidance technology.

DOI: 10.1504/IJRIS.2019.098055

International Journal of Reasoning-based Intelligent Systems, 2019 Vol.11 No.1, pp.65 - 75

Received: 22 Aug 2017
Accepted: 28 Nov 2017
Published online: 01 Mar 2019 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article