Li Auto and Tsinghua University Co-Develop AI System to Predict Pedestrian Safety, Cutting Head Injury Risk by 62.1%

Li Auto's safety team, in collaboration with Tsinghua University's Automotive Safety and Lightweighting team, has achieved a research breakthrough in the field of traffic safety. Their paper has been accepted for publication in the journal *Accident Analysis & Prevention*. The study proposes a "Predictive Integrated Pedestrian Safety System" that leverages AI-based short-term trajectory prediction technology to anticipate potential collisions approximately 200 milliseconds before impact and proactively activate protective measures. In complex scenarios—such as sudden pedestrian crossings or obstructed visibility—the system can preemptively deploy safety mechanisms, significantly reducing response latency compared to conventional pedestrian protection methods and lowering the risk of critical head injuries by up to 62.1%. Li Auto is currently advancing validation of this technology on real vehicle platforms to facilitate its practical implementation.