Group 1 - The report titled "2025 Insights on the Commercialization of L4 Intelligent Driving Scenarios in China" focuses on the progress of L4 intelligent driving commercialization in China, outlining the participants in the industry chain, core application scenarios, and development trends [1] - Key participants in the industry chain include technology companies like Baidu Apollo and Pony.ai, automotive manufacturers such as BAIC Group and GAC Group, communication and internet companies like Huawei and Tencent Cloud, hardware manufacturers like Hesai Technology and Horizon Robotics, and research institutions like Tsinghua University [1] - Core commercial scenarios for L4 intelligent driving have emerged in closed/semi-closed environments, such as port logistics and airport operations, with companies like JD Logistics and DeepBlue Technology leading the way [1] Group 2 - The report highlights the focus on multi-sensor fusion, vehicle-to-everything (V2X) collaboration, and AI model empowerment to enhance adaptability in complex environments [1] - Commercialization efforts emphasize cost control through large-scale applications and technological iterations to reduce hardware and operational costs, supported by policy incentives [1] - Future developments in L4 intelligent driving are expected to penetrate urban delivery and trunk logistics, with industry chain collaboration and scenario segmentation being key to deepening commercialization [1]

Group 1 - The core viewpoint of the article highlights that the company,拓邦股份, has successfully achieved large-scale supply of laser radar motors to leading new energy vehicle manufacturers, thereby consolidating its market share [1] - As the industry progresses towards L3+ level intelligent driving, the increase in laser radar installation rates is expected to open up new market opportunities [1] - The company is positioned to benefit from the rising penetration rate in the industry due to its automotive-grade technology and early customer advantages, which will further expand its market share in the intelligent driving supply chain [1]

Group 1 - The company has not yet directly engaged in Robotaxi operations but has accumulated rich experience in multi-dimensional data collection, ASLM risk control models, and Beidou free-flow scenarios [1] - The company aims to provide a collaborative foundation for related enterprises in intelligent driving model training, risk management, and demonstration applications [1] - The company plans to actively assess and expand cooperation opportunities with upstream and downstream partners in the Robotaxi industry chain based on national policy guidance and industry development trends [1]

Core Viewpoint - The article discusses the rapid growth of intelligent driving companies in China, highlighting the emergence of a market structure dominated by four key players: Huawei, Horizon, Momenta, and DJI. It emphasizes the significance of the "end-to-end" approach in intelligent driving systems, particularly through Horizon's advancements in their Horizon SuperDrive (HSD) system, which is positioned as the "Chinese version of FSD" [1][2][3]. Group 1: Market Trends and Growth - The penetration rate of L2 assisted driving functions is expected to reach 50% this year, surpassing the adoption rate of electric vehicles, which took 10 years to achieve the same level [1]. - Horizon has shipped over 10 million units of its intelligent vehicle chips and is set to mass-produce its HSD system [1][2]. Group 2: Technological Advancements - Horizon's "soft and hard integration" capability is highlighted as a strong competitive advantage, allowing for faster development cycles in the ICT industry [2]. - The transition to an "end-to-end" system is seen as a pivotal moment for the intelligent driving industry, moving away from traditional rule-based systems to a more streamlined, data-driven model [3][4]. Group 3: Product Development and Strategy - Horizon aims to avoid producing ordinary products by focusing on the end-to-end approach, which enhances user experience and safety [3][5]. - The company is concentrating on the passenger vehicle market and does not plan to diversify into other areas in the short term, emphasizing the importance of mastering this segment [9][12]. Group 4: Challenges and Future Directions - The article notes that achieving true autonomous driving remains a significant challenge, with many existing products lacking the necessary foundational capabilities [10][11]. - Key areas for improvement include increasing computational power, optimizing infrastructure, and standardizing sensors to support rapid iterations and scalability in the industry [11].

Core Viewpoint - Wanma Technology (300698.SZ) has commenced operations at its Changzhou Intelligent Computing Center, indicating a significant step in its technological capabilities and infrastructure development [1] Group 1 - The company's wholly-owned subsidiary, Shanghai Youka, is collaborating with NVIDIA on the application of NVAIE in the autonomous driving data closed-loop platform [1] - This collaboration aims to enhance the performance and technological iteration of the intelligent driving data closed-loop toolchain, which encompasses the entire process from data collection, processing, labeling, training, to simulation [1]

Core Viewpoint - Youjia Innovation (02431) has entered into a strategic partnership with Shenzhen Post and Eastern Bus to enhance resource utilization in the "logistics + public transport" sectors, leveraging each party's strengths in technology, support, and platform resources [1] Summary by Relevant Sections Strategic Partnership - The collaboration focuses on four main areas: integration of unmanned logistics vehicles with public transport scenarios, composite utilization of bus station spaces, expansion of cross-scenario business, and collaborative innovation in labor utilization [1] Resource and Technology Synergy - The board believes that the partnership will achieve complementary advantages in resources and technology, utilizing the company's extensive experience in L1–L4 autonomous driving technology and production, combined with Shenzhen Post's support and Eastern Bus's platform reuse mechanism [1] Cost Reduction and Efficiency - The partnership aims to effectively reduce redundant costs in warehousing and transportation, providing low-cost and highly adaptable solutions for the large-scale implementation of unmanned logistics in urban settings [1]

Core Viewpoint - Youjia Innovation (02431) has entered into a strategic partnership with Shenzhen Post and Eastern Bus to enhance resource utilization in the "logistics + public transport" sectors, leveraging each party's strengths in technology, support, and platform resources [1] Group 1: Strategic Partnership - The collaboration focuses on four main areas: integration of unmanned logistics vehicles with public transport scenarios, composite utilization of bus station spaces, expansion of cross-scenario business, and innovative labor collaboration [1] - Youjia Innovation will act as the technology and product supplier, while Shenzhen Post will provide support for scenario implementation, and Eastern Bus will serve as the unmanned vehicle platform provider [1] Group 2: Cost Efficiency and Innovation - The partnership aims to reduce redundant costs in warehousing and transportation by combining Youjia's advanced driving technology and production experience with the resources of Shenzhen Post and Eastern Bus [1] - This collaboration is expected to provide low-cost, highly adaptable solutions for the large-scale implementation of unmanned logistics in urban settings, creating a replicable model for cost reduction and efficiency improvement across the industry [1]

Core Insights - The article discusses the rapid development of large models and their integration into intelligent driving, highlighting the growing consensus in the industry regarding the commercial viability of L3 level intelligent driving by 2025 [1][2] - The introduction of the Visual-Language-Action (VLA) model is expected to create a comprehensive cognitive framework similar to human drivers, influencing the landscape of intelligent driving and embodied intelligence, while presenting significant market and capital opportunities [1][2] Group 1: Key Presentations and Insights - The event featured speeches from industry leaders, including Chen Zhongyi from Guotai Junan Securities and Wu Heng from SAIC Group, emphasizing the importance of L3 intelligent driving and embodied intelligence [3] - Zhu Feng, Chief Analyst at Guotai Junan, presented on the VLA model as a key to achieving L3 intelligent driving [3] - Yuan Yuji from Momenta discussed the company's data-driven approach and dual-product strategy for scalable autonomous driving solutions, including mass production of assisted driving and Robo taxi applications [4] Group 2: Technological Innovations - He Yihan from Che Lian Tian Xia highlighted the evolution of intelligent cockpit large models, focusing on redefining vehicle experiences through AI and optimizing multiple large language models for practical applications [5] - Liu Bin from Juefei Technology emphasized the importance of data loops in driving high-quality development in intelligent driving, providing customized data engines and services [5] - Zhou Enze from Al-Link showcased innovations in automotive intelligent cockpit technology, significantly reducing development costs for car manufacturers while enhancing user experience [6] Group 3: Industry Trends and Future Directions - Wang Panqu from Zero One Automotive discussed the transition to intelligent heavy trucks under the VLA framework, aiming to become a leading global transportation robotics company through vertical integration and innovative design [6] - A roundtable discussion involving industry, investment, and banking experts was held to explore the intersection of these sectors and their implications for the future of intelligent driving [6]

香港交易及結算所有限公司及香港聯合交易所有限公司對本公告的內容概不 負 責，對 其 準 確 性 或 完 整 性 亦 不 發 表 任 何 聲 明，並 明 確 表 示，概 不 對 因 本 公 告 全部或任何部分內容而產生或因依賴該等內容而引致的任何損失承擔任何責 任。 Minieye Technology Co., Ltd （於中華人民共和國註冊成立的股份有限公司） 深圳佑駕創新科技股份有限公司 （股份代號： 2431） 自願性公告 簽署無人物流與公交領域合作框架協議 根 據 合 作 協 議，三 方 將 以 各 自 核 心 資 源 與 渠 道 優 勢 為 基 石，以 業 務 協 同 為 紐 帶，深 度 挖 掘 跨 領 域 合 作 切 入 點，構 建 長 期 穩 定、互 惠 共 贏 的 戰 略 合 作 夥 伴 關 係。合 作 將 圍 繞 四 大 方 向 推 進： 1. 勞 動 用 工 協 同 創 新：建 立 跨 行 業 人 力 資 源 共 用 平 台，推 動 三 方 員 工 技 能 互 通 與 崗 位 柔 性 流 動，實 現 人 力 價 值 再 創 造。 – 1 – 簽訂合作協議的理由及裨益 三 方 在 相 關 領 ...

大模型发展如火如荼,将大模型进一步融合至智能驾驶中已成为产业共识, 而近年来政策正使得 L3 级智能驾驶落地商用渐成可能。在此背景下,视觉 - 语言 - 动作模型 (VLA) 应运而生，VLA 有望构建类似人类驾驶员的整体认知 框架,将影响智能笃驶、具身智能产业格局与技术发展路线,并带来巨大的市 场和资本机遇。 13:30-13:40 领导发言致辞 L3 高阶智驾专题暨 VLA 模型产业 自皮书及产业图谱发布 2025年9月4日(周四) 上海 - 国泰海通外滩金融广场 陈忠义 - 国泰海通证券副总裁、研究与机构业务委员 会总裁、研究所党委书记、政策和产业研究院院长 吴珩 - 上汽集团金融事业部 总经理、上汽集团金控管 理有限公司 总经理 13:40-14:00 通往 L3 智能驾驶与具身智能之钥 -- 视觉 - 语言 - 动作模型(VLA) 朱峰 - 国泰海通政策和产业研究院科技首席分析师 14:00-14:30 主题演讲 袁玉记 -Momenta 解决方案总监 Momenta (北京初速度科技有限公司)是全球领先自 动驾驶公司,致力于通过深度学习和人工智能技术实 现可规模化的自动驾驶解决方案。公司基于数据驱 ...