Core Viewpoint - The Heilongjiang Provincial Government has issued policies to promote high-quality development of e-commerce, aiming to integrate e-commerce with various sectors, innovate with new technologies like artificial intelligence, and enhance digital consumption levels to drive economic growth in Heilongjiang [1] Group 1: Support for E-commerce Entities - The policies encourage the cultivation of quality e-commerce entities, supporting new leading projects with total investments of 5 million yuan or more, providing interest subsidies on bank loans for infrastructure and equipment purchases, with a maximum subsidy of 2 million yuan per entity [1] - E-commerce platforms are supported for self-built or upgraded platforms, with interest subsidies for projects over 3 million yuan, capped at 200,000 yuan per entity [2] Group 2: Expansion of E-commerce Scale - E-commerce enterprises are encouraged to extend into production, improve supply chain collaboration, and broaden online sales channels, with a reward of 2% of the annual increase in physical online retail sales exceeding 10 million yuan, capped at 3 million yuan [2] Group 3: Cross-border E-commerce Support - Cross-border e-commerce enterprises are encouraged to expand import and export channels, receiving a 50% one-time subsidy for website construction and cloud service rental costs, with a maximum of 500,000 yuan per entity [3] - Support is also provided for e-commerce bases to upgrade using innovative technologies, with interest subsidies for projects over 1 million yuan, capped at 1 million yuan per entity [3] Group 4: Demonstration and Specialization - The policies promote the creation of smart e-commerce industrial bases, offering a one-time reward of 1 million yuan for national demonstration bases and 500,000 yuan for provincial innovation benchmark bases [4] - Support is given to establish specialty selection centers to integrate local quality products, with a one-time funding subsidy of 50% of actual investment for centers driving over 50 million yuan in annual online sales, capped at 1 million yuan [4] Group 5: Smart Logistics and Infrastructure - The policies support the intelligent transformation of e-commerce warehousing and logistics, with interest subsidies for projects over 5 million yuan, capped at 3 million yuan for warehousing construction or upgrades [5] - Cross-border warehousing network upgrades are encouraged, with interest subsidies for projects involving smart warehousing facilities, capped at 5 million yuan per entity [5] Group 6: Broader E-commerce Development - Additional measures are outlined to enhance e-commerce's role in empowering the real economy, accelerate public brand creation, strengthen the professional talent pool, and improve service support systems [6]

证券研究报告 行业研究 / 行业点评 2025 年 12 月 12 日 汽车 徐姝婧 S0820124090004 021-32229888-25517 xushujing@ajzq.com 一年内行业指数与沪深 300 指数对比走势： 资料来源：聚源数据，爱建证券研究所 相关研究 《车企年销量目标完成度分化——智能汽车数 据销量跟踪（二）》2025-12-03 《蔚来业绩创历史新高，25Q4 有望实现盈利 ——智能汽车系列报告（五）》2025-12-03 《理想召回事件短期扰动业绩，静待 2026 年 产品升级——智能汽车系列报告（六）》 2025-12-01 《小鹏 25Q3 亏损收窄，盈利拐点在望——智 能汽车系列报告（四）》2025-11-19 《零跑 25Q3 盈利持续改善，技术普惠驱动销 量高增——智能汽车系列报告（三）》 2025-11-18 吴迪 S0820525010001 021-32229888-25523 wudi@ajzq.com 联系人 行业及产业 奔驰 Robotaxi 于阿布扎比开启路测 ——Robotaxi 系列报告（一） 强于大市 事件概要：12 月 9 日，梅赛德斯-奔驰 ...

Core Viewpoint - Bosch, a leading international Tier 1 supplier, is heavily investing in both research and production lines for autonomous driving, aiming to keep pace with the rapid development in the domestic smart driving sector [2]. Group 1: Bosch's Investment and Research Directions - Bosch is focusing on two main areas: production and research, with significant resources allocated to end-to-end solutions [2]. - The company has recently recruited numerous technical experts to enhance its capabilities in autonomous driving [2]. - Bosch's research includes notable algorithms and projects, such as DGS and FlowDrive, which are aimed at improving autonomous driving technologies [2][4]. Group 2: Key Research Contributions - DGS (Dense Depth Regularization for LiDAR-free Urban Scene Reconstruction) is a framework that achieves high-quality geometric reconstruction and depth estimation without LiDAR, significantly reducing complexity and costs [5]. - FlowDrive introduces an energy flow field for end-to-end autonomous driving, enhancing safety and interpretability in trajectory generation [9]. - AnchDrive combines dynamic and static trajectory anchors to improve the efficiency and performance of trajectory generation in autonomous driving [13]. Group 3: Advanced Techniques and Frameworks - DiffSemanticFusion enhances the stability and semantic richness of online HD maps, improving trajectory prediction and planning tasks [16]. - IRL-VLA presents a closed-loop reinforcement learning framework that optimizes driving strategies without relying on high-fidelity simulators, achieving advanced performance metrics [19]. - SparseMeXT redefines online HD map construction using sparse representations, achieving superior accuracy and efficiency compared to existing methods [21]. Group 4: Data and Model Innovations - The Impromptu VLA dataset addresses performance issues in unstructured driving scenarios, providing a large-scale, high-quality resource for training VLA models [23]. - DiffVLA integrates vision-language models for enhanced decision-making in complex driving environments, demonstrating improved robustness and generalization [25]. - DINO-R1 introduces a novel training strategy for visual foundation models, significantly enhancing reasoning capabilities in visual prompt detection [27].

Core Viewpoint - The recent developments in the autonomous driving sector highlight the struggles of traditional automakers in establishing self-research capabilities, leading to a shift towards reliance on suppliers for technology solutions [1][11][14]. Group 1: Company Developments - Great Wall Motors launched the VLA intelligent driving model, showcasing its technological advancements in autonomous driving [1]. - Following the launch, its subsidiary, Haomo Zhixing, announced a complete shutdown, indicating the harsh realities of competition in the commercial sector [1]. - Haomo Zhixing's decline was foreshadowed by its failure to adapt to critical technological shifts, resulting in a significant reduction in workforce from 1500 to around 200 employees [1][3]. Group 2: Industry Trends - The trend among traditional automakers is moving away from self-research towards partnerships with suppliers, with companies like Chery and Great Wall Motors prioritizing manufacturing and sales over autonomous technology development [4][11]. - The autonomous driving industry has seen a significant decline in financing, dropping from 932 billion to 200 billion in three years, reflecting a cautious investment climate [17]. - The shift towards supplier solutions is becoming a standard practice, allowing automakers to focus on vehicle performance and brand identity rather than extensive in-house development [19]. Group 3: Competitive Landscape - New entrants like NIO, Xpeng, and Li Auto are positioned differently, focusing on building competitive advantages without the burden of traditional sales pressures [6]. - The failure of companies like Dazhuo Intelligent highlights the challenges faced by traditional automakers in balancing technological innovation with operational realities [10]. - The integration of external suppliers has become essential for traditional automakers to remain competitive in the rapidly evolving autonomous driving market [14][19].

Core Viewpoint - Xidi Intelligent Driving is set to launch its global IPO on December 19, aiming to become China's first publicly listed autonomous mining truck company, with an expected market capitalization exceeding HKD 11.5 billion [1] Group 1: IPO Details - Xidi Intelligent Driving plans to issue 5.40798 million H-shares, with 270,400 shares available in Hong Kong and 5.13758 million shares for international sale [1] - The IPO price is set at HKD 263 per share, with approximately 55% of the net proceeds allocated for R&D, 15% for enhancing commercialization capabilities, 20% for potential investments and acquisitions, and 10% for working capital [1] Group 2: Market Position and Performance - Xidi Intelligent Driving ranks sixth among all intelligent commercial vehicle companies in China, holding a market share of approximately 5.2% [2] - The company is projected to generate product sales revenue of approximately RMB 250 million in 2024, which accounts for about 16.8% of the market share [2] - Xidi Intelligent Driving's autonomous mining trucks have improved mining efficiency to 104% compared to manually operated trucks, making it the first and only company in China to achieve this [2] Group 3: Company Background and Financials - Founded by renowned scientist Li Zexiang, Xidi Intelligent Driving has completed eight rounds of financing since 2018, raising nearly RMB 1.5 billion from investors including Sequoia Capital and Lenovo Holdings [3] - The company reported revenues of RMB 31 million, RMB 133 million, and RMB 410 million for 2022, 2023, and projected for 2024, reflecting a compound annual growth rate of 263.1% [3] - Adjusted net losses decreased from RMB 159 million in 2022 to RMB 127 million in 2024, indicating a trend of narrowing losses [3]

Core Viewpoint - Rivian has announced its self-developed AI chip RAP1, a next-generation onboard computer, and a new AI model, aiming to replace Nvidia's solutions in its upcoming R2 model while introducing a subscription service called Autonomy+ [1][3][4] Group 1: Technology Development - Rivian plans to equip its upcoming R2 SUV with the Rivian Autonomy Processor 1 (RAP1) chip and a new lidar sensor, which are expected to enhance the company's autonomous driving capabilities [3] - The RAP1 chip utilizes multi-chip module technology with a memory bandwidth of 205GB per second, significantly improving performance compared to the current Nvidia system [3] - The new onboard computer, Autonomy Compute Module 3, can process 5 billion pixels per second, which is four times the performance of the existing Nvidia system used in Rivian vehicles [3] Group 2: Subscription Service - Rivian will launch the Autonomy+ subscription service in early 2026, priced at $2,500 for a one-time fee or starting at $49.99 per month, which is significantly lower than Tesla's FSD pricing [4] Group 3: Market Position and Strategy - Rivian's CEO RJ Scaringe emphasized the challenge of simultaneously reducing costs while enhancing performance, claiming that they have managed to lower vehicle costs by several hundred dollars while improving performance [6] - Despite a 25% increase in stock price this year, Rivian's stock is still down over 80% from its post-IPO peak [6] - Rivian aims to achieve Level 4 (L4) autonomous driving, allowing vehicles to operate without driver supervision, and plans to gradually roll out software updates starting in 2027 [11][12] Group 4: Competitive Landscape - Rivian's approach contrasts with Tesla's, as Rivian supports the use of lidar for environmental monitoring, while Tesla relies solely on camera-based systems [8][10] - Rivian's R2 model is set to begin production in the first half of 2026, but initial vehicles will not feature the new chip or lidar, limiting their autonomous capabilities [10] - Rivian's software system, Large Driving Model, will learn from driving behaviors to enhance the autonomous driving capabilities of older models equipped with Nvidia's Orin chip [13]

Core Viewpoint - The automotive industry is expected to continue the vehicle replacement policy until 2026, but Q1 2026 may face a demand overdraft period. Investors are advised to focus on globally competitive Chinese companies and embrace new industry trends [1][10]. Passenger Vehicle Market - The vehicle replacement policy is likely to continue, with Q1 2026 potentially being the worst period for the industry. Long-term investments should prioritize companies with overseas market potential [1][3]. - In the first ten months of 2025, China's wholesale passenger vehicle sales reached 24.17 million units, a year-on-year increase of 12.8%, with new energy vehicle sales at 12.18 million units, up 32% [2]. New Energy Vehicles - The purchase tax policy for new energy vehicles is set to decline in 2026, and while the vehicle replacement policy may continue, there will be changes compared to 2025. A demand overdraft period of approximately 3-4 months is anticipated [3]. - The total sales of new energy passenger vehicles in 2026 are expected to reach 1.811 million units, a year-on-year increase of 14.9% [3]. Autonomous Driving - 2025 is projected to be a year of accelerated penetration for intelligent driving, with high-speed NOA and urban NOA penetration rates reaching 16% and 14%, respectively [4]. - The intelligent driving industry is shifting from "function definition" to "data definition," with several trends emerging, including increased parameter counts in models and the importance of world models and reinforcement learning [4]. Humanoid Robots - Tesla's fourth chapter of its grand vision emphasizes bringing AI into the physical world, with humanoid robots being a core product. The Optimus V3.0 is expected to be released in Q1 2026 [5][6]. - The Chinese robot supply chain is well-established, and domestic companies are expected to play a significant role in the development of humanoid robots due to their manufacturing capabilities [6]. Commercial Vehicles - In the first ten months of 2025, commercial vehicle sales reached 3.472 million units, a year-on-year increase of 9%. The heavy truck segment is expected to benefit from the vehicle replacement policy [7]. - The export of heavy trucks in 2025 is projected to reach 274,000 units, with a year-on-year increase of 12% [7]. Two-Wheel Vehicles - The domestic market is undergoing a supply upgrade, with new national standards promoting industry normalization and leading to the exit of smaller brands. The market is expected to recover in the long term [8]. - Exports of large-displacement motorcycles increased by 59.1% in the first ten months of 2025, indicating strong overseas demand [8].

Core Viewpoint - Rivian Automotive has launched its first custom computer chip for autonomous driving, named Autonomy+, which is priced significantly lower than Tesla's offerings, marking a pivotal moment in the company's journey towards achieving Level 4 autonomous driving [2][5][6]. Group 1: Chip Development and Technology - Rivian's core technology roadmap focuses on transitioning to in-house developed chips designed for vision-centric physical AI, with the first generation being a custom 5nm processor [4]. - The new chip, capable of executing 16 trillion operations per second (TOPS) in a dual-chip configuration, integrates processing and storage into a single multi-chip module, enhancing efficiency and performance [4]. - Rivian claims its chip can process 5 billion pixels of camera data per second and utilizes RivLink technology for low-latency interconnectivity, which increases processing capabilities [4][6]. Group 2: Competitive Positioning - Rivian's move to develop its own chips aligns it with Tesla, which is also pursuing in-house chip development for autonomous vehicles, while other manufacturers increasingly rely on Nvidia [5][6]. - The company plans to integrate LiDAR into its upcoming R2 model to improve redundancy and real-time driving performance, contrasting with Tesla's approach of not using LiDAR [5][8]. Group 3: Product Offerings and Pricing - The Autonomy+ driving assistance suite is priced at $2,500 for a one-time payment or $49.99 per month, significantly lower than Tesla's $8,000 full self-driving system [7][8]. - Rivian aims to launch a "no-look" driving feature by 2026, leveraging its advancements in chip technology and AI models [8]. Group 4: Long-term Goals and Market Strategy - Rivian's long-term goal is to achieve Level 4 autonomous driving, allowing vehicles to operate without human intervention under specific conditions [6]. - The company emphasizes deep vertical integration as a key strategy to reduce costs by eliminating supplier margins and customizing components according to its needs [6].

Core Viewpoint - The company, Mainline Technology, has submitted an application for a mainboard listing on the Hong Kong Stock Exchange, with Cathay Pacific Securities acting as the sole sponsor [1] Company Overview - Mainline Technology is a Level 4 autonomous driving truck and solution provider, focusing on its self-developed autonomous driving system [1] - The company has created an integrated product ecosystem consisting of AiTruck (smart trucks), AiBox (smart terminals), and AiCloud (smart cloud services) [1] Business Strategy - The company's solutions are strategically deployed in three core business scenarios: Trunk Port (logistics hub), Trunk Pilot (highway logistics), and Trunk City (urban traffic) [1] - As of December 5, 2025, the company has delivered a total of 830 AiTrucks and 349 AiBoxes, with intention orders for 821 AiTrucks and 920 AiBoxes [1] Market Position - According to Frost & Sullivan, Mainline Technology is the largest Level 4 autonomous driving truck and solution provider in China for closed road scenarios, based on projected product sales revenue for 2024 [1] Industry Outlook - The market for autonomous driving solutions in commercial vehicles in China is rapidly developing, with the closed road scenario market expected to reach 26.2 billion RMB by 2030, while the open road scenario market is projected to reach 160.6 billion RMB during the same period, which is anticipated to be the core growth engine for the next decade [1]

智通财经APP获悉，中信证券发布研究报告称，预计2026年汽车行业以旧换新政策延续的概率较大，但 Q1行业或仍将面临一段时间的需求透支期。建议投资人聚焦具有全球竞争力的中国企业，全面拥抱产 业新趋势。2026年上半年有望跑赢的投资标的包括：1)出海景气延续、出海盈利弹性大的乘用车和商用 车龙头企业;2)自动驾驶加速渗透带来的头部智驾公司、上游产业链、L4公司的投资机会;3)人形机器人 的产业趋势，将继续为板块提供业绩和估值的双重驱动，建议聚焦特斯拉等人形机器人企业的上游核心 零部件公司。 中信证券主要观点如下： 乘用车总量：以旧换新政策有望延续，2026Q1或是行业最差时间，优先选择出海品种进行长期布局。 根据中汽协数据，2025年1-10月，我国乘用车批发销量为2417万辆，同比+12.8%，新能源乘用车批发销 量为1218万辆，同比+32%，渗透率为50.4%。今年乘用车同比增长超预期，主要受益于以旧换新补贴政 策的延续、海外多个市场出口销量的增长。根据商务部发布的1-11月补贴总量数据，预计2025全年总补 贴申请合计将达1240万份，预计总补贴金额将达1650亿元，对2025年的实际销量增量拉动可能达 ...