期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2017
卷号：95
期号：23
页码：6454
出版社：Journal of Theoretical and Applied
摘要：Congestion remains as an important issue to date, and it is continuously investigated by experts in order to find solutions in the process of information dissemination to road users. The existing research was more concerned on detecting vehicles and road rather than congestion, thus there has been no clear definition of congestion offerred. What really happened was that vehicle accumulation will always be referred to as congestion, although it is not necessarily the case, thus the information about congestion condition becomes inaccurate. Moreover, the existing research mainly viewed the vehicle speed aspect as the basis of congestion level estimation. Other researchers used GPS and probe detectors, but there has been much limitation despite such equipment. On average, the research carried out was only focused on the number of vehicles at a given frame as the basis for density/congestion calculation. In fact, congestion is also determined by road density and vehicle speed in a given period. Therefore, it is necessary to find a way that can display the information about road condition and congestion in a factual, timely manner. In this paper, we present two novel methods in the context of congestion detection employing information about road density and vehicle speed. Road density defines the extent to which road area is occupied by vehicles, while vehicle speed sees how fast vehicles pass a given frame. Thus, the information of both helps define the levels of vehicle congestion on a road. Based on the experiment conducted on an in-city road, this method was found to be accurate in defining the levels of congestion, starting from light, jam to heavy-jam traffic. To corroborate the argumentation of the congestion conditions, calculation using Fuzzy model was conducted given that the congestion levels are not measurable in an exact manner (light (macet ringan), jam (macet sedang), heavy-jam (macet berat) and total traffic gridlock (macet total)), thus the information obtained is more accurate. The method developed does not require high cost, yet it is quite effective in presenting congestion information in a quick, real-time, accurate way.
