End-to-end intelligent driving for all scenarios and the nationwide NOA that has already been rolled out are both navigation on autopilot. What is the difference in the driving experience between the two?
The accumulation from the fully rolled-out mapless NOA serves as the foundation for end-to-end capabilities. End-to-end all-scenario intelligent driving enables the activation of smart driving features right from a standstill, bridging driving and parking. It offers a higher ceiling for user experience, delivering more coherent and consistent performance.
End-to-end technology incorporates user driving data for training. Building upon the mapless foundation, it provides more human-like trajectories, more efficient navigation, and safer driving. With the support of the VLM (Vision-Language Model), it can handle even more complex scenarios.
More human-like driving: The trajectory during driving is smoother and more natural, with a significant reduction in sudden acceleration and hard braking. It can intelligently anticipate and plan the optimal driving path, making more reasonable driving decisions in advance.
More efficient navigation: Starting right from the parking space, it is ready to use as soon as you get in the car. It features stronger negotiation capabilities, flexibly handling oncoming traffic, pedestrians, and static obstacles. It can navigate through ETC lanes and barrier checkpoints, achieving a seamless point-to-point journey from parking space to parking space.
Safer driving: The end-to-end model's flexible multi-dimensional evaluation mechanism conducts real-time safety assessments of trajectories, embedding the safety baseline into its core DNA. It reacts faster to complex road conditions involving mixed traffic with two-wheelers and pedestrians.
VLM handles more complex scenarios: It analyzes road conditions in real time, identifying and alerting the driver to potholes, construction zones, T-junctions, and more.