" 用一次几乎没有退路的选择，实现一场「务实浪漫主义」的胜 利。 " 作者丨李雨晨 编辑丨 林觉民 2016年，大疆内部诞生了一个神秘项目，编号BR1609，意为这一年的第九个预研项目，这就是卓驭的前 身。 2024年9月27日，大疆车载事业部从大疆的天空之城总部搬出，正式更名卓驭。独立之时，卓驭CEO沈劭 劼的心里"预研"了分拆后的各种可能。 他坦言，"那一刻的心情，就像是《三体》里人类变成太空人类的过程。敌人太强大，内部资源不足，残 酷到人类有五艘船，但最终的资源只够两艘生存，必须得灭掉三艘，我的心境有点像这个样子。" 当时的卓驭，面对的是一个智驾"修罗场"。为了拿下客户，各家智驾Tier1使出浑身解数，用驻场培训、 算法白盒等方式，试图拿到几款销量不明的车型。 与拆分同步进行的是卓驭当时正在做的一轮融资。当时，卓驭融资15亿，但实际只到账了三分之一。上千 人的公司，账上的钱只够维持小几个月。 2025年的倒数第二个工作日，卓驭举办了为数不多的一次大型品牌活动，这也是其九年技术长征的阶段性 总结。 从2016年跨界切入智能驾驶，到2024年10月14日删库重炼、全面转型端到端的决绝，再到如今转向多元 移动 ...

来源：环球网 "过去大家比的是'有没有NOA'，未来两三年，比的是'能不能全天候稳定运行'。"他认为，真正的门槛 不再是单一的感知或规划算法，这些技术在行业内已经阶段性趋同。真正的挑战在于系统工程能力—— 能否跨城市、跨天气、跨道路风格，提供一致性的体验。 2025年，汽车行业的AI叙事正在发生微妙的转向。过去几年，车企们在发布会上争相展示"城市NOA开 城数量"、"算力堆叠规模"，仿佛参数就是一切。但在资深汽车行业从业者龙骁显看来，这场关于智能 驾驶的"军备竞赛"正在进入一个新的拐点：从"功能有无"的炫技，转向"系统可靠"的交付。当技术的光 环褪去，消费者真正愿意为AI买单的理由是什么？行业分析认为，智能驾驶正面临从技术验证到商业 成功的关键跨越。 龙骁显强调："这要求车企不能只做一个'算法工作室'，而必须进化为'系统大厂+软件公司'。"他指出， 从演示级表现到规模化可交付，中间隔着容错机制、冗余设计和极端工况的安全架构。"赢家，属于那 些能把'炫技'变成'日常'的企业。" 龙骁显表示："如果说前两年是智驾的'战国时代'，大家比的是谁的武器更花哨；那么未来两三年，就 是'结硬寨、打呆仗'的阶段。"他敏锐地捕 ...

今年前三个季度，国内20万元以上乘用车市场份额占比30%，13万元以下市场份额则高达50%，但后者 多数车型尚未配备城区辅助驾驶功能。这一广阔的蓝海市场，正吸引着地平线、Momenta等智驾厂商加 速布局，全力抢占市场先机。 今年4月，地平线正式推出基于征程6系列芯片的城区辅助驾驶解决方案——HSD（Horizon SuperDrive）。尽管并非该赛道的先行者，但地平线已快速迈入大规模量产阶段。11月，随着星途ET5 正式上市，地平线的HSD解决方案同步实现量产；另一款搭载该方案的车型深蓝L06也于同期发售。两 款车型上市短短两周后，地平线HSD的激活量便突破12000辆，量产落地成效显著。 除了推出全新的解决方案，地平线还通过生态拓展加速市场渗透。12月初的地平线技术生态大会上，公 司公布了两大生态推进举措：一是拓展生态合作模式，新增算法服务模式"HSD Together"，并已与日本 电装、大众的合资公司CARIZON（酷睿程）、HCT（智驾大陆）达成合作；二是引入更多生态合作伙 伴，元戎启行、卓驭等企业已加入其生态体系。 缺乏芯片研发能力的算法公司、软硬研发实力薄弱的车企，正纷纷向地平线聚拢。地平线接 ...

Core Insights - The Ministry of Industry and Information Technology has approved the commercial operation of L3 autonomous driving for the first time in China, allowing vehicles to operate under specific conditions with the system taking over driving tasks [1] - The approval includes two models: Changan Deep Blue SL03 and Arcfox Alpha S6, marking a significant step towards the mass production of L3 autonomous vehicles by 2026 [1] - The responsibility for accidents occurring while the system is activated will primarily fall on the car manufacturers, emphasizing the importance of accountability in this new phase of autonomous driving [1] Industry Developments - Major automotive companies in China, including Huawei, Chery, and GAC Group, are targeting the implementation of L3 conditional autonomous driving by 2025, with several already obtaining testing licenses [4][5] - Companies like XPeng Motors and Chery have announced plans to launch L3 autonomous vehicles, with XPeng aiming for L4 capabilities by 2026 [4] - The L3 level is seen as a crucial transition from "assisted driving" to "fully autonomous driving," with L4 expected to allow vehicles to operate without human intervention in designated areas [1][4] Technological Advancements - The automotive industry is experiencing a shift towards integrating AI and advanced technologies into autonomous driving systems, with companies developing models that enhance perception, planning, and control [9][12] - The introduction of VLA (Visual Language Action) models is expected to significantly improve the capabilities of autonomous driving systems, providing better scene understanding and decision-making [9][15] - The competition among automakers is intensifying, with a focus on developing proprietary technologies that enhance vehicle performance and safety, particularly in complex driving scenarios [17][18] Future Outlook - The approval of L3 autonomous driving is viewed as a pivotal moment in the evolution of transportation, setting the stage for ongoing exploration and innovation in the field [19] - The industry is expected to continue evolving, with a focus on balancing self-research and collaboration to maintain technological leadership while managing costs [18][19] - As the market for autonomous vehicles grows, the emphasis will shift from merely achieving autonomous capabilities to ensuring the safety and reliability of these systems in real-world conditions [17][19]

AI给智驾的赋能触及瓶颈了？ "明星高管"苏菁再次"大胆开麦"引发圈内热议。 "大家最近都很嗨，一段式（端到端）搞定了，各种乱七八糟的概念都跑出来了，智驾是不是会进入一 个大发展时代？"近日，在2025地平线技术生态大会，地平线副总裁兼首席架构师苏箐表示："我 要'浇'点冷水。" 他认为，这一代深度学习技术可能碰到天花板了，所以很大概率在未来三年内，智驾的发展是在现有的 系统上做极致优化，而不是理论内核重构，所以"大家别太嗨，又进入苦日子的阶段了"。 公开资料显示，苏箐曾担任华为车BU智能驾驶产品部部长，负责华为自动驾驶系统方案。在华为车BU 时，苏菁就因为"敢说"在智驾圈广为人知，曾发表"基础是计算机，车是计算机控制的外设"的观点，并 锐评特斯拉智驾事故"讲难听点就是'杀人'"。 在此次地平线技术生态大会上，苏菁回顾了过去几年其在智驾行业的历程，这也映射了整个智驾行业近 年来的发展。2022年他准备加入地平线时，心态是"不想再做自动驾驶了"，原因一是太痛苦，二是看不 到希望，"那个时代的自动驾驶看似不错，但与真正人类司机相比差距极大。" 当时的智驾技术范式主要是规则驱动。分水岭出现在2024年初特斯拉FSD ...

Core Viewpoint - The article discusses the technological anxiety in intelligent driving, particularly among mid-tier manufacturers, and highlights the anticipated growth in demand for end-to-end (E2E) and VLA (Vision-Language-Action) technologies in the coming year [2]. Group 1: Industry Trends - The mass production of cutting-edge technologies like end-to-end systems is expected to begin next year, with L2 technology becoming more standardized and moving towards lower-tier markets [2]. - The total sales of passenger vehicles priced above 200,000 are around 7 million, but leading new forces account for less than one-third of this, indicating a slow adoption of end-to-end mass production models [2]. - The maturity of end-to-end technology is seen as a precursor to larger-scale production, with the advancement of L3 regulations prompting urgent upgrades among mid-tier manufacturers [2]. Group 2: Recruitment and Training - There is a growing demand for positions related to end-to-end and VLA technologies, as many professionals are seeking to quickly learn these advanced skills [3]. - The article mentions the launch of specialized courses aimed at practical applications of end-to-end and VLA technologies, designed for individuals already working in the field [3][6]. - The courses will cover various modules, including navigation information application, reinforcement learning optimization, and production experiences related to diffusion and autoregressive models [3][6]. Group 3: Course Details - The end-to-end production course will focus on practical implementation, including seven major practical applications, making it suitable for those looking to advance their careers [3][6]. - The VLA course will cover foundational algorithms and theories, including BEV perception and large language models, with practical projects based on diffusion models and VLA algorithms [6][11]. - The instructors for these courses are experienced professionals from top-tier companies and academic institutions, ensuring a high-quality learning experience [5][8][13].

卓驭 沈劭劼 一家智驾供应商如何从规则崩塌中练出 "神奇的模型"。 文 丨 宋玮 郭瑞婵 如果不离开大疆，青年教授沈劭劼大概率不会遇到 "喝红的还是喝白的""要不要跪着满足所有需求""账 上只够发三个月工资怎么办" 这种完全不属于工程世界的难题。 但这恰恰是卓驭这家公司的起点。 2024 年 9 月 27 日，沈劭劼带着 1000 多名员工搬离了大疆深圳总部 "天空之城"。这一天，他正式告别 被视为工程天堂的 "温室"，走向资源有限、竞争残酷的现实世界。 大疆出身，既是光环，也是枷锁。 卓驭一出生就站在聚光灯下，被放进 "地大华魔" 的坐标里。但也长期活在巨人投影之下——大疆以产 品为中心，而卓驭必须学会以客户为中心；大疆内部天然的资源池，让分拆后的卓驭在销售、客户关系 等能力上几乎从零起步。工程底子很强，商业肌肉几乎空白，这是它矛盾又特殊的起点。 卓驭就像一个天赋极高、但缺乏社会经验的学霸，需要在市场里重新习得生存之道。 今年 11 月，它从大疆绝对控股的团队，变成由一汽、大疆和管理层共同治理的 "无实际控制人" 公司。 国资委以 "链长" 方式推动一汽入局，新一轮融资超 36 亿元、估值 120 亿元。这 ...

Core Viewpoint - The article discusses the decline of the autonomous driving supplier, Haomo Technology, highlighting its operational challenges, loss of key partnerships, and the impact of its reliance on Great Wall Motors [4][22]. Company Overview - Haomo Technology, incubated by Great Wall Motors in 2019, was once a leading player in the autonomous driving sector, primarily supplying technology for Great Wall's various vehicle brands [4][6]. - The company had a peak workforce of around 800 employees, focusing on autonomous driving technology, but has since dwindled to fewer than 300 employees [4][6]. Key Events - In 2023, Haomo faced significant setbacks, including the loss of its exclusive partnership with Great Wall Motors, which shifted to another supplier, Yuanrong Qixing, for its new models [5][10]. - Despite retaining some contracts with Great Wall for lower-tier models, Haomo's overall market position weakened as it became one of several suppliers for other major automakers like Hyundai, Toyota, and BMW [8][10]. Financial and Operational Challenges - Haomo's financial health has deteriorated, with a reported valuation of approximately 900 million yuan in 2024, showing limited growth compared to its previous valuation of 1 billion USD in 2021 [21][22]. - The company has undergone five rounds of financing, raising about 1.5 billion yuan, but has struggled to convert its technology into cash flow due to operational inefficiencies and a lack of market traction [21][22]. Technological Development - Haomo's technology strategy has been criticized for lagging behind competitors, particularly in the transition to more advanced autonomous driving systems that do not rely on high-definition maps [14][17]. - The company initially focused on a "heavy perception, light mapping" approach but failed to adapt quickly to industry shifts towards end-to-end models, which integrate perception, prediction, and planning into a single system [15][17]. Market Position and Future Outlook - The article suggests that Haomo's close ties with Great Wall Motors have limited its ability to establish deeper partnerships with other automakers, hindering its growth potential [22]. - As the industry moves towards more autonomous solutions, Haomo's inability to innovate and adapt may lead to further decline, raising questions about its long-term viability [22].

Core Viewpoint - The news highlights the decline of Haomo Technology, a smart driving supplier, which has faced significant challenges leading to a halt in operations and a drastic reduction in workforce, primarily due to its inability to keep pace with technological advancements and market demands [1][2][19]. Company Overview - Haomo Technology was incubated by Great Wall Motors in 2019 and initially thrived as a smart driving star company, primarily supplying smart driving systems for various Great Wall brands [1][19]. - The company had a peak workforce of nearly 800 employees, focusing on smart driving technology development, but has since dwindled to less than 300 employees [1][2]. Market Dynamics - In 2023, Haomo lost a significant contract with Great Wall's Weipai brand to a competitor, Yuanrong Qixing, due to delays in the mass production of its urban NOA (Navigation on Autopilot) feature [2][7]. - Despite retaining some contracts with Great Wall and other automakers like Beijing Hyundai, Toyota, and BMW, Haomo is not the sole supplier for these companies, limiting its market position [5][6]. Technological Challenges - Haomo's reliance on Qualcomm chips has been a double-edged sword, providing some partnership opportunities but also limiting its computational capabilities compared to competitors using NVIDIA platforms [8][10]. - The company has struggled with the timely adaptation of its technology roadmap, particularly in transitioning to more advanced autonomous driving solutions, which has hindered its competitive edge [10][14]. Financial Situation - Haomo has undergone five rounds of public financing, raising approximately 1.5 billion yuan, but its valuation has only increased modestly from 1 billion USD in 2021 to around 9 billion yuan in 2024 [17][19]. - The company’s IPO plans were halted due to internal decisions, reflecting a lack of confidence in the current market conditions and its financial performance [16][17]. Strategic Missteps - Haomo's strong dependency on Great Wall Motors has limited its ability to forge deeper partnerships with other automakers, which is critical in the competitive landscape of smart driving technology [19]. - The company has faced difficulties in converting its technological advancements into tangible cash flow, leading to operational challenges and ultimately a decision to halt operations [16][19].

Core Insights - Xiaomi Auto officially launched the Xiaomi HAD Enhanced Version at the Guangzhou Auto Show, showcasing advancements in smart driving technology and talent acquisition in the AI field [1] - The company plans to invest over 7 billion yuan in AI research and development by 2025, with a current team of 1,800 experts, including 108 PhDs [1] Technical Developments - The new Xiaomi HAD Enhanced Version is built on a foundation of 10 million clips and incorporates reinforcement learning algorithms and world models to enhance driving performance [1] - The world model technology allows the system to simulate various scenarios, including extreme weather and complex road conditions, transitioning from a "rule-driven" to a "learning-driven" approach [1] User Experience Enhancements - The updated version focuses on optimizing longitudinal and lateral control experiences, particularly in scenarios like lane merging, reducing unnecessary deceleration and hard braking [2] - Significant upgrades to active safety features include the introduction of the AES emergency steering assist function, which can automatically change lanes to avoid collisions at speeds between 80 km/h and 135 km/h [2] Safety Features Expansion - The forward AEB (Automatic Emergency Braking) range has been expanded to 1 km/h to 135 km/h, with new capabilities to recognize various obstacles [2] - The backward AEB covers reversing scenarios from 1 km/h to 30 km/h, with a focus on balancing sensitivity to ensure accurate stopping while minimizing false triggers [2] Software Updates - The driving updates will be included in the Xiaomi HyperOS 1.11.0 version, with rollout times varying by model due to review progress [2]