news

한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina

IT Home reported on July 18 that according to 36Kr Auto, Ideal Auto recently established an entity organization for "end-to-end autonomous driving".A total of more than 200 peopleAt the same time, some members of other teams in the company are also flexibly supporting the project.

According to reports, the Ideal Smart Driving team is mainly divided into two groups: algorithm research and development and mass production research and development.Team of about 800 peopleThe main force of "end-to-end" R&D is deployed in the algorithm R&D group, which is led by Jia Peng, the head of Ideal Intelligent Driving Technology R&D.

Algorithm R&D GroupIt consists of groups such as perception algorithm, behavioral intelligence, and cognitive intelligence. "Perception algorithm" is led by Chang Li; "Behavioral intelligence" includes end-to-end architecture, end-to-end model, control model, etc., which is in charge of Yang Yi; "Cognitive intelligence" includes cognitive model, cloud model, etc., which is actually in charge of Zhan Kun.

Mass production R&D teamWang Jiajia, the head of mass production, is in charge of the project, which includes smart driving, smart parking, smart safety, etc. Jia Peng and Wang Jiajia both report to Lang Xianpeng, the head of Ideal Intelligent Driving.

IT Home noticed that in June this year, Li Xiang, Chairman and CEO of Ideal Auto, revealed at the 2024 China Automotive Chongqing Forum that in the third quarter, an end-to-end + VLM (visual language model) supervised autonomous driving system trained with 3 million Clips (pre-trained models) will be launched to test users.From the end of this year to the beginning of next year, will launch a supervised autonomous driving system trained with more than 10 million clips.

During the Ideal Smart Driving Summer Conference held on July 5, Ideal Auto announced thatEnd-to-end + visual language model early bird plan officially launched, claiming to make cars smarter and more human-like.

Ideal Auto said that the advantages of the end-to-end model lie in "efficient transmission" and "efficient computing": the end-to-end model is an integrated model, and information is transmitted within the model, which has a higher upper limit.The actions and decisions of the entire system that users can feel are "more human"The integrated model can complete reasoning in the GPU at one time, and the end-to-end latency is lower. Users can feel the coordination between "eyes" and "hands", and the vehicle's actions respond in a timely manner.