Application of Interval Type-2 Fuzzy Sets in Unmanned Vehicle Visual Guidance

Hong Mo,Xuanming Zhao,Fei-Yue Wang
DOI: https://doi.org/10.1007/s40815-019-00680-4
IF: 4.085
2019-07-08
International Journal of Fuzzy Systems
Abstract:In the paper, a novel unmanned vehicle visual guidance is proposed by applying type-2 fuzzy comprehensive evaluation and time-varying universe. Unlike processing all image information in the traditional methods, the new guidance describes the visual focus of human drivers in a more practical way by using interval type-2 fuzzy sets, which can address the problem of linguistic ambiguity and data noise superior to conventional fuzzy sets. Based on the sets above, this method applies the fuzzy comprehensive evaluation method for the computer to select the appropriate visual focus and merges time-varying universe to establish a fuzzy visual guidance rule library which is close to human driving. During the process of image processing, it can assist with selecting the more reasonable visual focus depending on the rule library. The examples indicate that the method can help to make efforts at cost reduction of visual processing, which shorts the response time and reaction distance greatly.
computer science, information systems,automation & control systems, artificial intelligence
What problem does this paper attempt to address?