按2024年收入计算，在全球所有L4级无人驾驶公司中排名第一。 本文为IPO早知道原创 作者｜ Stone Jin 微信公众号｜ipozaozhidao 据 IPO早知道消息， 易控智驾科技股份有限公司 （以下简称 " 易控智驾 "）于2025年6月25日正 式向港交所递交招股说明书，拟主板挂牌上市，海通国际担任独家保荐人。 根据弗若斯特沙利文的资料 ， 按 2024年收入 计算， 易控智驾在全球所有 L4级无人驾驶公司中排 名第一 ；截 至 2024年12月31日及 2025年6月18日 活跃无人驾驶矿卡数 计算 ，易控智驾是全球 最大的矿区无人驾驶解决方案提供商 ；同时， 易控智驾 还是 全球首家 、 也是目前唯一一家实现 1,000+台活跃无人驾驶矿卡的公司 ，截至 2025年6月18日 已部署一支由 超 1,400辆活跃无人驾 驶矿卡组成的车队 。 截至 2025年6月18日 ，易控智驾拥有 11家终端客户集团，技术已部署在包括国家能源集团、国家 电投、特变电工、紫金矿业、首钢集团 、 宝武集团等公司运营的 24个矿场 。 值得注意的是， 2022年至2024年连续三个年度，易控智驾在终端客户集团中保 ...

Group 1 - The Nasdaq China Golden Dragon Index (HXC) has included Pony.ai, marking a significant recognition of China's autonomous driving technology in the global capital market [2] - The index now comprises 73 Chinese companies, with Pony.ai being the only representative of cutting-edge technology as the first and only L4 autonomous driving company [2] - This inclusion is expected to attract hundreds of millions of dollars in incremental funds from passive investments such as ETFs and hedge funds, enhancing liquidity and valuation [2] Group 2 - Pony.ai's seventh-generation autonomous driving system has reduced hardware costs by 70%, with specific reductions of 80% in onboard computing units and 68% in lidar costs, achieved through partnerships with major automotive manufacturers [3] - The company anticipates a 200% year-on-year increase in revenue from its Robotaxi business in 2024, with a more than 20% growth in registered users, indicating the emergence of scale effects [3] - Once the fleet size exceeds 1,000 vehicles, the company expects to achieve a dynamic balance between operating costs and revenue, initiating a positive cycle of profitability [3] Group 3 - Following the index inclusion, the Nasdaq China Golden Dragon Index rose by 3.3%, with Pony.ai's stock surging 16%, reflecting market optimism towards the autonomous driving sector [4] - This trend indicates a shift in investment logic from "model innovation" to "hard technology-driven" approaches within the Chinese concept stock market [4] Group 4 - Pony.ai is expanding its technology solutions globally, having established a strategic partnership with the Dubai Roads and Transport Authority to advance the commercial operation of fully autonomous Robotaxis [7] - The company has also initiated road tests in cities like Seoul and Luxembourg, collaborating with Singapore's ComfortDelGro to develop transportation services [7] - Middle Eastern sovereign wealth funds have invested in Pony.ai, aligning its technology output with local smart city strategies [7] Group 5 - The integration of technology, capital, and market dynamics is becoming clearer for Pony.ai as it approaches mass production of its seventh-generation Robotaxi, driving the revolution in transportation [5]

Could Pony AI be the next big autonomous driving stock? Wall Street analysts think so -- here's why.Pony AI (PONY 16.73%) is leading the race to deploy fully autonomous robotaxis -- and its recent advances could propel the stock to new highs. Discover how partnerships with Uber, Tencent, and Toyota position Pony AI for breakout growth, why analysts are bullish, and what this $1 trillion disruptor might deliver next.Stock prices used were the market prices of June 16, 2025. The video was published on June 23 ...

近日，纳斯达克中国金龙指数对其成分股进行新一轮调整，中国Robotaxi公司小马智行正式纳入其中。 金龙指数是中概股投资标的风向标，纳入该指数意味着以小马智行代表的中国自动驾驶科技进入主流投 资视野，吸引ETF基金、对冲基金、长线投资者的投资，公司股票流动性和资本市场地位将进一步提 升。 ...

作者 | LiangWang 编辑 | 自动驾驶之心 点击下方 卡片 ，关注" 自动驾驶之心 "公众号 戳我-> 领取 自动驾驶近15个 方向 学习 路线 >>点击进入→ 自动驾驶之心 『4D标注』技术交流群 本文只做学术分享，如有侵权，联系删文 近几年随着深度学习技术的发展，基于数据驱动的算法方案在自动驾驶/机器人领域逐渐成为主流，因此算法对数据的要求也越来越大。区别于传统单帧标注，基 于高精点云地图的4D标注方案能够有效减少标注成本并提高数据真值质量。 4D标注中的4D是指三维空间+时间维度，4D数据能够映射到任意时刻得到单帧真值用于模型训练，区别于大范围高精地图生产，4D标注只关注一小片区域的静态 和动态元素。然而如何生成标注所需底图是其中的一个关键环节，针对不同的标注需求，通常需要实现"单趟建图"，"多躺建图"和"重定位"等关键技术，在场景上 还需要支持有GNSS的行车场景和无GNSS的泊车场景。 LSD (LiDAR SLAM & Detection) 是一个开源的面向自动驾驶/机器人的环境感知算法框架，能够完成数据采集回放、多传感器标定、SLAM建图定位和障碍物检测 等多种感知任务。 本文将详细介 ...

Market Overview - The global Robotaxi market is projected to reach $1.95 billion in 2024 and $43.76 billion by 2030, with a forecasted market size of 834.9 billion yuan by 2030 according to Tianfeng Securities [1] Competitive Landscape - Small Horse Intelligent (小马智行) and WeRide (文远知行) are the leading players in the autonomous driving sector, with significant differences in their operational strategies and technology focus [2][5] - Small Horse Intelligent emphasizes redundancy and safety in its technology, while WeRide focuses on cost optimization and a diversified product matrix [7][9] - Both companies have raised approximately $1.3 billion in funding, indicating strong investor interest in the autonomous driving sector [9] Financial Performance - Small Horse Intelligent's revenue from 2022 to 2024 was $68.39 million, $71.90 million, and $75.03 million, totaling approximately $215 million [16] - WeRide's revenue during the same period was 528 million yuan, 402 million yuan, and 250 million yuan, totaling approximately 1.18 billion yuan, indicating a significant decline in revenue [16] - As of the end of 2024, WeRide's total assets were 7.694 billion yuan, with a net asset growth of 331.52%, while Small Horse Intelligent's total assets were $1.051 billion, reflecting a 40.70% increase [18][19] Strategic Initiatives - Small Horse Intelligent is focusing on the Chinese market, with plans to expand its Robotaxi fleet to 1,000 vehicles by the end of 2025, while WeRide is pursuing a global expansion strategy [21] - Both companies are engaged in a competitive race for the title of "Robotaxi first stock," with WeRide successfully listing on NASDAQ first, achieving a market cap of $4.491 billion on its debut [12] Technology and Innovation - Small Horse Intelligent's technology emphasizes a dual approach of Robotaxi and Robotruck, utilizing a multi-sensor fusion strategy for its seventh-generation Robotaxi [7] - WeRide has developed a diverse product matrix that includes Robotaxi, Robobus, Robovan, and Robosweeper, showcasing its adaptability across various scenarios [9] Market Dynamics - The competition between Small Horse Intelligent and WeRide is characterized by a focus on "technical depth" versus "scene breadth," indicating a long-term strategic battle in the autonomous driving space [22]

论文标题 ： Structured Labeling Enables Faster Vision-Language Models for End-to-End Autonomous Driving 论文作者： Hao Jiang, Chuan Hu, Yukang Shi, Yuan He, Ke Wang, Xi Zhang, Zhipeng Zhang 论文链接： https://www.arxiv.org/pdf/2506.05442 作者 | Hao Jiang 来源 | 深蓝AI 点击下方 卡片 ，关注" 自动驾驶之心 "公众号 戳我-> 领取 自动驾驶近15个 方向 学习 路线 >>点击进入→ 自动驾驶之心 『端到端自动驾驶』技术交流群 本文只做学术分享，如有侵权，联系删文 引言 最近将类人的推理能力融入到端到端自动驾驶系统中已经成为了一个前沿的研究领域。其中，基于 视觉语言模型的方法已经吸引了来自工业界和学术界的广泛关注。 现有的VLM训练范式严重依赖带有自由格式的文本标注数据集 ，如图1(a)所示。虽然这些描述 能够 捕捉丰富的语义信息，但 由于两种结构不同但是表达相近的句子会增加模型在学习任 ...

Core Insights - The article presents MMTL-UniAD, a unified framework for multimodal and multi-task learning in assistive driving perception, which aims to enhance the performance of advanced driver-assistance systems (ADAS) by simultaneously recognizing driver behavior, emotions, traffic environment, and vehicle actions [1][5][26]. Group 1: Introduction and Background - Advanced driver-assistance systems (ADAS) have significantly improved driving safety over the past decade, yet approximately 1.35 million people die in traffic accidents annually, with over 65% of these incidents linked to abnormal driver psychological or physiological states [3]. - Current research often focuses on single tasks, such as driver behavior or emotion recognition, neglecting the inherent connections between these tasks, which limits the potential for cross-task learning [4][3]. Group 2: Framework and Methodology - MMTL-UniAD employs a multimodal approach to achieve synchronized recognition of driver behavior, emotions, traffic environment, and vehicle actions, addressing the challenge of negative transfer in multi-task learning [5][26]. - The framework incorporates two core components: a multi-axis region attention network (MARNet) and a dual-branch multimodal embedding module, which effectively extract task-shared and task-specific features [5][26]. Group 3: Experimental Results - MMTL-UniAD outperforms existing state-of-the-art methods across multiple tasks, achieving performance improvements of 4.10% to 12.09% in the mAcc metric on the AIDE dataset [18][26]. - The framework demonstrates superior accuracy in driver behavior recognition and vehicle behavior recognition, with increases of 4.64% and 3.62%, respectively [18][26]. Group 4: Ablation Studies - Ablation experiments indicate that joint training of driver state tasks and traffic environment tasks enhances feature sharing, significantly improving task recognition accuracy [22][26]. - The results confirm that the interdependence of tasks in MMTL-UniAD contributes to overall performance and generalization capabilities [22][26].

而自从端到端和大语言LLM横空出世以来，大规模无监督的预训练 + 高质量数据集做具体任务的微调， 可能也会成为量产感知算法下一阶段需要发力的方向。同时数 据的联合标注也是当下各家训练模型的实际刚需，以往分开标注的范式不再适合智能驾驶的算法发展需求。今天自动驾驶之心就和大家一起分享下4D数据的标注流 程： 最复杂的当属动态障碍物的自动标注，涉及四个大的模块： 而为了尽可能的提升3D检测的性能，业内使用最多的还是点云3D目标检测或者LV融合的方法： 得到离线单帧的3D检测结果后，需要利用跟踪把多帧结果串联起来，但当下跟踪也面临诸多的实际问题： 离线3D目标检测； 离线跟踪； 后处理优化； 传感器遮挡优化； 点击下方 卡片 ，关注" 自动驾驶之心 "公众号 戳我-> 领取 自动驾驶近15个 方向 学习 路线 千万级4D标注方案应该怎么做？ 智能驾驶算法的开发已经到了深水区，各家都投入了大量的精力去做量产落地。其中一块最关键的就是如何高效的完成4D数据标注。无论是3D动态目标、OCC还是静 态标注。 相比于车端的感知算法，自动标注系统更像是一个不同模块组成的系统， 充分利用离线的算力和时序信息，才能得到更好的感知结果 ...

点击下方 卡片 ，关注" 自动驾驶之心 "公众号 戳我-> 领取 自动驾驶近15个 方向 学习 路线 摘要 对于自动驾驶车辆而言，在复杂环境中安全导航依赖于应对广泛且多样化的罕见驾驶场景的能力。基于仿 真和场景的测试已成为自动驾驶系统开发与验证的关键方法。传统场景生成依赖基于规则的系统、知识驱 动模型和数据驱动的合成方法，但这些方法往往生成的场景多样性有限，且难以生成真实的高风险安全关 键场景。随着基础模型（Foundation Models）的出现——一种预训练的通用人工智能模型——开发者能够 处理异构输入（例如自然语言、传感器数据、高清地图和控制指令），从而实现对复杂驾驶场景的合成与 解析。本文围绕基础模型在自动驾驶场景生成与分析中的应用（截至2025年5月）开展综述研究。本综述提 出了一个统一分类体系，涵盖大语言模型（LLMs）、视觉-语言模型（VLMs）、多模态大型语言模型 （MLLMs）、扩散模型（DMs）和世界模型（WMs）在自动驾驶场景生成与分析中的应用。此外，我们回 顾了相关方法论、开源数据集、仿真平台和基准测试挑战，并探讨了针对场景生成与分析的专用评估指 标。最后，本文总结了当前面临的开放性 ...