TU-TESL

• 22

Since October this year, kuweather has entered into cooperation with well-known domestic and foreign automobile enterprises, such as, automobile, etc., to provide specialized and customized meteorological solutions for its Internet of vehicles and business. Kuweather is a professional weather service provider who specializes in travel scenarios. On the basis of providing various fine weather services, rwis pavement weather forecast system has been independently developed, which can provide driving conditions forecast of water, ice, snow, freezing rain, road temperature and other driving conditions of kilometer level roads in the next 7 days, as well as Visibility Forecast along the road. The service has been successfully applied to Baidu map and other mainstream map navigation, and at the same time, it works together with the car enterprise partners to create a high-precision, high-frequency refreshing car regulation level service. In the application scenario of Internet of vehicles, through rwis pavement weather forecast, the management center can make macro-control, and the driver can intuitively obtain the road status in front of the driving route as the basis for adjusting the driving route and regulating the speed, so as to ensure the arrival time and driving safety. At the same time, kuweather also provides users with refined weather forecast for more than 70000 scenic spots around the world and refined weather services for more than 10000 transportation hubs across the country, including personalized services such as weather phenomenon, air quality, sun protection index and flower and maple appreciation, so as to enhance the driving experience of car owners. In recent years, the Internet of vehicles and automatic driving have gradually become the focus of the industry; however, the interference of severe weather on visual algorithm and hardware has always been an important problem for the commercialization of automatic driving. Compared with the high cost and long cycle of sensor iteration, it is the best choice for the auto driving vehicle factory to directly introduce the existing capabilities of professional meteorological service providers. Further from the perspective of automatic driving scenario, sensor sensing, algorithm processing, data upload, driving computer algorithm response to physical braking all need a certain time. Compared with the real-time data detected by sensors, it is more reasonable to use the road state prediction as the auxiliary input of the automatic driving algorithm.