Nozomi Koyama, Soichiro Yokoyama, T. Yamashita, H. Kawamura, Kiyotaka Takeda, Makoto Yokogawa
{"title":"Recognition of snow condition using a convolutional neural network and control of road-heating systems","authors":"Nozomi Koyama, Soichiro Yokoyama, T. Yamashita, H. Kawamura, Kiyotaka Takeda, Makoto Yokogawa","doi":"10.1109/IESYS.2017.8233573","DOIUrl":null,"url":null,"abstract":"The conventional control method for road-heating systems, which is controlled mainly by snowfall, is the feedforward control. Therefore, the road-heating systems often operate in spite of no snow on the parking lots. In this paper, we introduce the method of feedback control for road-heating systems that detects snow condition on the parking lots, and turns road-heating systems on and off properly. To recognize the snow condition, we adopt images for recognition using convolutional neural networks.","PeriodicalId":429982,"journal":{"name":"2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 21st Asia Pacific Symposium on Intelligent and Evolutionary Systems (IES)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IESYS.2017.8233573","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
The conventional control method for road-heating systems, which is controlled mainly by snowfall, is the feedforward control. Therefore, the road-heating systems often operate in spite of no snow on the parking lots. In this paper, we introduce the method of feedback control for road-heating systems that detects snow condition on the parking lots, and turns road-heating systems on and off properly. To recognize the snow condition, we adopt images for recognition using convolutional neural networks.