直击进博会｜地平线与大众进入“后联姻时代” 联手造“芯”打造下一代“车脑”

Core Insights - The collaboration between Volkswagen Group (China) and Horizon Robotics marks a significant step in the development of autonomous driving technology in China, focusing on local design and development of system-level computing solutions [1][4][5] Group 1: Joint Venture and Investment - Volkswagen's software company CARIAD and Horizon Robotics have established a joint venture named CARIZON to design and develop system-level chips in China, with an investment of approximately $200 million [1][4] - The new system-on-chip (SoC) is expected to achieve mass production within 3 to 5 years, boasting a computing power of 500-700 TOPS, which is more than double that of the current leading Nvidia Orin-X chip [1][2] Group 2: Strategic Goals and Local Adaptation - The strategic goal of the joint venture is to create a "smart brain" for the next generation of intelligent connected vehicles, enhancing real-time decision-making and safety in complex driving conditions [2][3] - Volkswagen aims to localize core technology development in China, recognizing the market's critical role in future innovations and adapting to local driving habits and traffic conditions [4][6] Group 3: Phased Development and Future Plans - The first advanced driver assistance system solution from CARIZON is set to enter mass production in 2025, marking the completion of the first phase of their collaboration [3][5] - The second phase will focus on enhancing Volkswagen's local R&D capabilities in intelligent driving, aiming for large-scale production and commercial deployment of integrated intelligent driving solutions [3][7] Group 4: Industry Context and Competitive Landscape - The partnership reflects a shift in strategy among multinational automotive companies in China, moving from reliance on overseas R&D to integrating local innovation capabilities [6][8] - The collaboration is seen as a response to the rapid innovation pace of domestic brands in the fields of new energy and intelligent driving, aiming to shorten the iteration cycle of intelligent driving technologies [6][7]