Core Viewpoint - The tightening of regulations in the intelligent driving industry is seen as beneficial for clarifying product definitions and safety levels, leading to more transparent marketing practices [1][2]. Group 1: Regulatory Changes - The Ministry of Industry and Information Technology (MIIT) has emphasized that car manufacturers must avoid exaggerated and false advertising, ensuring clear communication of product capabilities [1][2]. - The MIIT has mandated the standardization of terminology, discouraging the use of terms like "autonomous driving" and "intelligent driving," favoring terms like "L2 assisted driving" instead [1][2]. Group 2: Industry Response - Following the MIIT's directives, companies have rapidly adjusted their marketing materials, with some removing the term "intelligent driving" from presentations before major events like the Shanghai Auto Show [1][2]. - The China Automotive Industry Association and the China Automotive Engineering Society have called for the promotion of safe application of combined driving assistance functions and the elimination of misleading marketing practices [2]. Group 3: Market Outlook - The intelligent driving industry is expected to undergo a淘汰赛 (elimination race) over the next one to two years, with a prediction that only a few key players will remain by 2027 [2]. - The competitive landscape is characterized by strong economies of scale and first-mover advantages, with a forecast that China could produce two to three leading companies in the global market [2]. Group 4: Technological Trends - Leading domestic intelligent driving companies are adopting a soft and hard integration approach, which is viewed as a correct direction for achieving optimal performance and cost efficiency [3]. - The focus on soft and hard integration is expected to enhance user experience and pricing strategies in the intelligent driving sector [3].

Core Viewpoint - The article emphasizes that Horizon Robotics is not merely a chip company but aims to become a software company, focusing on integrated hardware-software systems for autonomous driving, particularly in urban environments [2][4][5]. Group 1: Company Background and Development - Horizon Robotics was founded in 2015, and its founder, Yu Kai, recognized the nearing end of the mobile internet boom, leading the company to focus on robotics chips instead of algorithms [7]. - The company faced significant challenges in 2019, including a reduction in workforce by one-third, but pivoted to focus solely on the automotive sector [7][8]. - Collaborations with major automotive manufacturers like Changan and Li Auto have been pivotal in establishing trust and achieving market penetration [8][10]. Group 2: Market Position and Achievements - Horizon Robotics has achieved over 8 million units of front-mounted production and has a market share of 33.97% in the L2 autonomous driving computing solutions market for domestic brands [13]. - The company claims that one in three smart vehicles is equipped with its technology, projecting rapid growth towards 10 million units [14]. Group 3: Future Vision and Strategy - Horizon Robotics aims to redefine the essence of smart driving, focusing on both functional and emotional value, similar to how smartphones evolved beyond basic communication [15]. - The company plans to transition from L2 to L3 and L5 autonomous driving capabilities, requiring significant computational power and extensive real-world data [18][21]. - The introduction of the Horizon Smart Driving (HSD) system, featuring the Journey 6P chip, is set to enhance urban autonomous driving capabilities and is expected to be launched in 2025 [24][28]. Group 4: Technological Approach and Innovation - The HSD system utilizes a one-piece end-to-end technology architecture, ensuring high performance and efficiency in data processing [24]. - Horizon Robotics emphasizes the importance of a robust software-hardware integration strategy, akin to successful models like Apple and NVIDIA, to maintain a competitive edge [23]. Group 5: Market Dynamics and Competitive Landscape - The article discusses the competitive landscape, highlighting that while other companies may have launched similar technologies earlier, Horizon Robotics prioritizes product quality over timing [30]. - The company is aware of the potential pitfalls of technological advancements leading to homogenization in the market and focuses on building a sustainable competitive advantage through consistent R&D efforts [20].

Core Viewpoint - The recent humanoid robot marathon in Beijing, with a completion rate of only 30%, highlights the current limitations of humanoid robots in real-world scenarios, suggesting that the industry is at a critical juncture for innovation and opportunity [1][5][31]. Group 1: Event Overview - The marathon featured 20 humanoid robot teams, with only 6 completing the race, showcasing various technical failures such as falling, overheating joints, and battery issues [1][8][11]. - The course was intentionally designed with complex terrains, including sharp turns and steep slopes, which posed significant challenges for robots accustomed to controlled environments [11][15]. Group 2: Industry Implications - The low completion rate is viewed not as a failure but as a necessary step in the evolution of the humanoid robotics industry, akin to early automotive challenges that preceded a transportation revolution [5][31]. - The event serves as a wake-up call for entrepreneurs to shift focus from performance-driven development to solving real-world problems, indicating a potential shift from "performance economy" to "real-world application" [20][21][31]. Group 3: Technological Limitations - The marathon exposed three fundamental limitations of humanoid robots: inadequate three-dimensional environmental perception, lack of adaptive control in movement, and energy management issues leading to frequent battery changes [18][19][20]. - These limitations suggest that current innovations are insufficient and that a new approach is needed to achieve significant advancements in humanoid robotics [20][31]. Group 4: Market Opportunities - The article identifies three key areas for potential market growth: industrial inspection and flexible manufacturing, disaster rescue in hazardous environments, and home care services for the elderly [24][28]. - Entrepreneurs are encouraged to explore upstream components, midstream platforms, and downstream solutions to create a comprehensive ecosystem that addresses real-world needs [25][26]. Group 5: Future Directions - The concept of "second curve innovation" is emphasized, where the industry must move beyond superficial performance to address genuine challenges, thus unlocking new market opportunities [20][31]. - The article concludes that the current state of humanoid robotics is not the end but rather the beginning of a transformative journey, urging stakeholders to prepare for the next phase of innovation in the AI landscape [31][33].

2025年春日的北京亦庄，一场特殊的马拉松赛事引发全网热议：20支人形机器人队伍参赛，最终仅6支完 赛，完赛率30%。 机器人跌倒、换电、关节过热等 "翻车"场景，让公众从短视频里的"炫酷幻想"跌回现实：原来，那些能跳 舞、翻跟头的"钢铁舞者"，在真实复杂的开放场景中，竟如此脆弱。 业界难免疑惑，是人形机器人跌倒在马拉松赛道？还是整个人形机器人赛道就此跌倒？ 哲科思维往往是集体群嘲中的一丝冷静，如果用善友教授的创新理论来看，这场看似 "失败"的赛事，恰恰 是行业第二曲线的关键一役。 这些看似狼狈的 "跌倒"，实则是产业进化的必经之路。 正如汽车诞生初期跑不过马车，却预示着交通革命的前夜。 对于创业者，这是 AI的黎明，机会近在眼前；对于普通人，这是一次预见未来生活的窗口。 这场只有 30%完赛率的机器人马拉松 ， 并不是闹剧，而是意味着： 新一波赚钱机会，终于来 了。 这将是一个千亿级别的赚钱机会。 人形机器人跌倒在赛道？还是让整个赛道跌倒？ 这场马拉松在大众眼里，仿佛只是场娱乐活动。 当人类选手在赛道上肆意挥洒汗水时， 20支机器人队伍却交出了一份"惨不忍睹"的答卷：仅有6支队伍完 赛，完赛率30%。 天工 ...

Core Viewpoint - The article discusses the rising trend of intelligent driving technology in the automotive industry, emphasizing the importance of safety and rationality in its development [1][21]. Group 1: Industry Developments - The Ministry of Industry and Information Technology (MIIT) of China has initiated measures to ensure the safety of intelligent driving systems, requiring car manufacturers to conduct thorough testing and avoid exaggerated claims [2]. - Horizon Robotics launched its L2 urban driving assistance system, Horizon HSD, which is powered by the high-performance Chengcheng 6P chip, capable of 560 TOPS [3][5]. Group 2: Product Features and Capabilities - The HSD system covers various driving scenarios, including urban areas, highways, and parking lots, and can perform complex maneuvers such as three-point turns and navigating narrow roads [6]. - The HSD system is designed to balance efficiency and safety, learning from the driver's habits while outperforming novice drivers in certain situations [6][8]. Group 3: Market Position and Growth - Horizon Robotics has established itself as a leader in the L2 driving assistance market, achieving a market share of 33.97% and delivering over 8 million units [11][12]. - The company aims to deliver over 10 million units by 2025, capitalizing on the expected increase in the penetration rate of L2 driving assistance systems in new vehicles [11]. Group 4: Financial Performance and Future Outlook - Horizon Robotics reported a revenue of 2.384 billion yuan in 2024, a 53.6% increase year-on-year, with a gross profit margin of 77.3% [16]. - The company anticipates achieving profitability by 2027, marking a significant milestone in its growth trajectory [17]. Group 5: Research and Development - Horizon Robotics plans to invest approximately 3 billion yuan in R&D in 2024, focusing on advancing its algorithms and technology [19]. - The company has received the SO 8800AI functional safety certification, enhancing its credibility in the global market [19]. Group 6: Strategic Partnerships and Ecosystem - Horizon Robotics collaborates with over 40 automotive manufacturers, including major players like Chery and Volkswagen, to expand its market presence [11][20]. - The company is building an ecosystem that integrates chips, algorithms, and developer partnerships to strengthen its competitive position [12].

Core Viewpoint - The Chinese automotive industry is emphasizing the need for companies to eliminate false advertising and excessive marketing, ensuring the safety of intelligent connected vehicles, especially in light of recent incidents like the Xiaomi SU7 fire accident [1] Group 1: Industry Initiatives - The China Automotive Industry Association and the China Automotive Engineering Society issued a proposal to promote self-discipline among companies and ensure quality commitments and safety responsibilities [1][2] - The proposal calls for companies to clarify and rectify any false information they have released and to take on both corporate and social responsibilities [2] Group 2: Key Aspects of the Proposal - Companies are urged to fulfill their product quality safety responsibilities by enhancing their capabilities, establishing robust quality management systems, and ensuring timely and understandable information prompts for users [2] - Marketing practices must adhere to the standards set by GB/T 40429—2021, avoiding misleading statements and ensuring scientific naming of functions to prevent misuse by drivers [2] - A user notification mechanism should be established to inform drivers about their responsibilities, system information, proper usage, and emergency procedures, ensuring a clear understanding of the differences between driving assistance and autonomous driving [2] Group 3: Industry Perspectives - The Ministry of Industry and Information Technology emphasized the importance of conducting thorough testing and validation of driving assistance systems, highlighting the need for companies to take responsibility for product consistency and safety [3] - Industry leaders, such as Horizon's CEO, noted that tighter regulations would positively impact the industry by promoting a focus on safety rather than merely pursuing autonomous driving [3] - NIO's CEO stressed that safety should be a top priority and that all stakeholders, including government and society, must work together to enhance traffic safety [4]

【文/观察者网 张家栋 编辑/高莘】 4月18日，地平线正式发布L2城区辅助驾驶系统HSD以及征程6P计算方案。 地平线L2城区辅助驾驶系统HSD 文中图片均来自地平线官网 一周前工信部对于"智能辅助驾驶"的监管力度升级，覆盖的不仅是整车厂商，作为软硬件技术输出方的 智驾供应商同样位列其中。 而作为一家乘着智驾浪潮一战成名，且刚刚上市不久的智驾明星企业，地平线与行业的另一位领跑者华 为一样，同样在经历智驾的信任危机。 如何在对"智驾"的反思中，为智驾赋予新的活力，正在成为地平线思考的问题。在这场发布会上，余凯 给出了4个观点： 第一点：智能驾驶向用户提供的是功能价值，而并非类似智能座舱提供的情绪价值。余凯认为，"智能 驾驶如同智能汽车的'基带'，是重要的基础设施。"这意味着，智能驾驶不再具有品牌的个性，而是对 于整个汽车作为移动空间，给用户提供安全、便捷、舒适等标准化的"功能价值"。 第四点：智驾技术在发展过程中还远未及L3以上场景。在这一点上，余凯再度强调了其对于智驾的观 点，即坚持长期主义，用耐得住寂寞来成就客户。面对全行业对于L3的高呼，余凯称，"L3/L4/L5的前 提是足够好的L2全场景辅助驾驶。" ...

Core Insights - The demand for computing power for Level 4 (L4) autonomous driving is projected to reach 5000 TOPS by 2030-2035 [1] - Level 3 (L3) autonomous driving requires extensive deployment of Level 2 (L2) systems and validation through real-world data, with a computing power requirement of 500-1000 TOPS expected in 2-3 years [1] - The automotive industry is increasingly focusing on democratizing intelligent driving technologies, emphasizing the importance of high computing capabilities for survival and competitiveness [1] Group 1 - The foundation for L3, L4, and L5 autonomous driving relies on effective full-scene assisted driving systems [1] - Achieving L3 requires a significant number of L2 full-scene assisted driving system vehicles and thorough validation of real-world data [1] - Companies must either develop their own technologies or collaborate with third parties to keep pace with advancing technical requirements [1]

【导读】地平线正式宣布高算力芯片征程6P和6H流片量产，并推出高性能L2地平线城区辅助驾驶系统HSD 地平线创始人兼CEO余凯称，地平线想做机器人时代的计算平台公司，即"芯片+操作系统"。 "当征程6P成为标配时，预计单车ASP（平均售价）会提升至少十倍，公司2025年的市场占有率将会随之进一步提升。"余 凯对中国基金报记者表示，"公司的护城河来自市场竞争地位，这是我们目前最关心的。对于盈利转正，我们一点都不担 心。预计最快在2027年，地平线将迎来盈利拐点。" 4月18日，地平线股价以5.47港元报收，最新市值为722亿港元。 进一步推进"软硬一体" 4月18日晚，地平线创始人兼CEO余凯在上海滴水湖正式宣布，高算力芯片征程6P和6H流片量产，并推出高性能L2地平线 城区辅助驾驶系统 Horizon Super Drive（以下简称HSD）。 其中，征程6P的算力达到560 TOPS，相对于市面主流的大算力芯片，征程6P的性能有两位数提升，但单位成本却并没有大 幅提升。 HSD搭载征程6P，采用一段式端到端技术架构，是国内首个软硬结合全栈开发的L2城区辅助驾驶系统。 4月18日，在公司成立即将满十周年之际， ...

地平线上，日出东方。地平线创始人兼CEO余凯一出场就是豪情万丈。上海临港，在这个最早看见日出的地方，余凯仿佛看到中国汽车工业如一轮红日 冉冉升起。 2025年4月18日，以"征程所向，远超想象"为主题的2025地平线年度产品发布会隆重举行。这家中国智驾领域的"隐形冠军"，首次对外完整披露其十年 发展历程，并推出L2城区辅助驾驶系统——地平线HSD，构建体验"类人"、用户"信任"的城区辅助驾驶新解。同时，地平线与奇瑞达成HSD首个量产合作， 以全栈实力助力智能化量产迈向新征程。 发布会现场，余凯以近两小时的演讲，回溯公司从"科学家创业"到"企业经营者"的蜕变，同时用硬核技术参数、量产数据与用户反馈，展现了其在智能 驾驶领域的深度布局与行业野心。 十年磨剑：从"耐得寂寞"到"生态之王" 地平线的故事始于2015年。彼时，余凯刚刚离开百度，创立地平线，目标直指"为未来无处不在的机器人提供计算平台"，而计算平台正是"芯片+操作系 统"。创业初期，公司选择了一条人迹罕至的路径——聚焦人工智能芯片，而非当时热门的移动互联网或云计算。"在没有竞争的地方竞争"成为其核心战 略。 "耐得寂寞才是真正的英雄主义，成就客户是自我成 ...