Xiaomi Lei Jun: SU7 is expected to upgrade end-to-end assisted driving HAD in late July, a new 10 million Clips version

Passionategeekz On June 25, Xiaomi Auto announced on June 23 that the entire series of Xiaomi YU7 will be equipped with 10 million Clips version end-to-end assisted driving. Officials said that more new developments will be announced on the evening of June 26.

Passionategeekz noticed that Xiaomi founder, chairman and CEO Lei Jun posted a post this morning to share the recent progress of Xiaomi’s end-to-end assisted driving HAD: the new 10 million Clips version, the entire series of Xiaomi YU7 is equipped with it as soon as it was released from the factory;Xiaomi SU7 is expected to be upgraded in late July. In addition, he said, “Compared with the current version,The improvement is still very obvious。”

According to previous reports, at the launch of the Guangzhou Auto Show in November last year, Lei Jun announced that Xiaomi Smart Drive has entered a new stage and will be named Hyper Autonomous Driving (HAD for short)— Xiaomi super intelligent driving. Xiaomi Auto mentioned in the Q&A session that in terms of end-to-end capabilities, Xiaomi will soon end-to-end driving and parking scenes, with a higher experience limit and more coherent and consistent performance.

The accumulation of NOA without a picture is the end-to-end basis. It introduces user driving data training end-to-end. On the basis of without a picture, the trajectory is more personified, the traffic is more efficient, and the driving is safer. The VLM visual language model can cope with more complex scenarios after being supported.

• Driving is more anthropomorphic: the trajectory during driving is smoother and more natural, with significantly reduced acceleration and sudden braking. It can intelligently predict and plan the best driving path and make more reasonable driving decisions in advance.

• Passage is more efficient: the parking space is started, and it can be used as soon as it gets on the bus. It has stronger gaming ability and flexibly responds to incoming cars, pedestrians, and static obstacles. It can achieve ETC and gate junctions. There is no breakpoint throughout the journey from the parking space to the parking space.

• Safer driving: The free multi-dimensional evaluation mechanism of end-to-end models conducts real-time safety assessment of trajectories, engraving the safety bottom line in the genes, and two-wheelers and pedestrians confuse complex road conditions and respond faster.

• VLM deals with more complex scenarios: real-time analysis of road conditions, identify potholes, construction scenarios, T-junctions, etc. and prompt the driver.

Advertising statement: The external redirect links (including, not limited to, hyperlinks, QR codes, passwords, etc.) contained in the article are used to convey more information and save selection time. The results are for reference only. All articles from Passionategeekz include this statement.