Hokkaido University and Software Development Vitz (Nagoya City, President Hiroyuki Hattori) have started the development and demonstration experiment of automatic operation for snowy and cold regions in the eastern part of Tomakomai City, Hokkaido.Development of autonomous driving and demonstration experiments are underway all over the country, but it is likely to attract attention because the development of technology for cold regions where it is difficult to check the situation around the vehicle due to snowfall is delayed.
According to Hokkaido University, five companies, Yamaha Motor, Ark System Solutions, Aisin Comcruise, AW Software, and Atelier, also participated in the project.A research support project for small and medium-sized enterprises of the Ministry of Economy, Trade and Industry's Hokkaido Bureau of Economy, Trade and Industry, aiming to realize autonomous driving in cold regions by developing self-position estimation technology on snowy roads by utilizing new image processing technology and AI (artificial intelligence) ..The automatic driving location was set to the eastern part of the city with the cooperation of Tomakomai City and others.
Development competition is underway around the world for autonomous driving as a technology of the near future.Even in Japan, universities and IT companies are entering into demonstration experiments in various places due to the government's voice.Especially in depopulated areas where the population is declining and the population is aging, public transportation such as railroads and buses is weakening, and there are great expectations for the practical application of automatic driving due to the frequent occurrence of accidents caused by the elderly driving.
However, in the current automatic driving technology, the case of estimating one's own position by utilizing high-precision spatial information is the mainstream.In places with a lot of snow, signs and road boundaries are covered with snow, and self-position estimation using high-precision spatial information does not work.The challenge was to establish a self-position estimation technology that would work even in places with heavy snowfall.
