Home / Metal News / Li Auto rolls out ‘end-to-end + VLM’ smart driving system via OTA

Li Auto rolls out ‘end-to-end + VLM’ smart driving system via OTA

iconOct 23, 2024 17:17
Source:gasgoo
On October 23, Li Auto announced the full-scale rollout of its first End-to-End + Visual Language Model (VLM) intelligent driving system.

Beijing (Gasgoo)- On October 23, Li Auto announced the full-scale rollout of its first "End-to-End + Visual Language Model (VLM)" intelligent driving system. This upgrade, part of the OTA 6.4 update, will be available for owners of the Li MEGA, L9, L8, L7, and L6 AD Max models.

Li Auto's new intelligent driving system aims to enhance safety and ease of use, mimicking the driving behavior of experienced human drivers in various driving scenarios. Notable features include the ability to activate end-to-end driving directly from Park (P) mode when starting from a roadside spot, smoother U-turns with human-like precision, and more assertive and calculated decision-making at roundabouts.

The system also improves performance in congested traffic, with enhanced lane-changing capabilities and smoother bypass. At complex intersections, it selects the optimal lane with precision, and it can efficiently navigate special lanes, such as bus lanes and variable lanes, while adhering to traffic regulations.

Powered by the VLM, the system excels in recognizing and handling challenging scenarios such as potholes, speed bumps, construction zones, and night driving without streetlights. This results in smoother, safer driving experiences with advanced obstacle anticipation.

Li Auto began its early adopter program for the "End-to-End + VLM" system in July 2024, highlighting the efficiency of the end-to-end model in rapid processing and low-latency responses. The system's intelligent capabilities are expected to improve over time as it continuously learns and adapts to new driving conditions, making it a better and safer driving companion.

For queries, please contact William Gu at williamgu@smm.cn

For more information on how to access our research reports, please email service.en@smm.cn

SMM Events & Webinars

All