LiDAR has high upstream barriers, and domestic manufacturers give priority to opening up midstream breakthroughs​

To join the "Smart Car Experts Lidar Industry Exchange Group", please add 18512119620 (same as WeChat), note company-name-position to join the group

China's autonomous driving penetration rate is increasing, and vehicle-mounted lidar has broad prospects. According to Sullivan's forecast, the global lidar market is expected to reach US$13.54 billion in 2025, and the Chinese lidar market is expected to reach US$4.31 billion, accounting for 31.8%. We believe that with the rapid development of artificial intelligence and autonomous driving technology, China will continue to expand its market share and become the world's major lidar sales market. With the continuous breakthroughs and upgrades in intelligent technology, driven by the expansion of driverless fleets and the increased penetration of lidar applications in advanced assisted driving, the lidar market is expected to expand rapidly. my country's automotive lidar market size will reach 460 million yuan in 2021, and the market size is expected to reach 5.47 billion yuan in 2025, achieving a compound annual growth rate of 85.8%.

The upstream of lidar mainly includes four modules: transmission, reception, information processing and scanning. The transmitting optical system at the transmitting end includes a beam controller, excitation source, and laser. The cost of different light sources varies greatly. The transmitting end is an important cost component of lidar, accounting for 30%-50%. The receiving end converts the optical signal into an electrical signal, mainly including a photodetector and a processing chip, and finally amplifies the signal and uses the chip for processing. Beam control requires a scanning system, and different scanning methods will have different structural designs.

Specifically, the upstream of the industrial chain can be divided into four parts. Overall, overseas manufacturers occupy the main market share, but in recent years, domestic substitution has quickly occupied the market with cost advantages:

➢ Lasers and detectors: mainly overseas, including ams-OSRAM, Lumentum, HAMAMATSU, II-VI, etc .; domestic suppliers include Zonghuixinguang, Ripple Optoelectronics, Huaxin Semiconductor, Yangtze Huaxin, etc. The main suppliers of fiber lasers include overseas Luminar , Lumibird, IPG Optoelectronics, O-Net, and domestic Leishen Intelligence. In terms of detectors, overseas suppliers mainly include ams Osram, ON Semiconductor, Firstsensor, HAMAMATSU, Sony, etc.; domestic suppliers mainly include Quantum Core Integration, Lingming Photon, Core Vision Microelectronics, Guangxun Technology, Fushi Technology, etc.

➢ FPGA: FPGA is usually used as the main control chip of lidar. The mainstream overseas suppliers include Xilinx (acquired by AMD), Altera (acquired by Intel), Lattice, etc.; the main domestic suppliers include Unisoc and Zhidao Crystal Microelectronics, Fudan Microelectronics, Anlu Technology, Gowin Semiconductor, etc. In terms of performance, domestic development still lags behind overseas, but the logical resource scale and high-speed interface performance of domestic products can also meet the needs of lidar. Although FPGA is currently the mainstream solution, as the demand for performance and overall system increases, more integrated SoC may be widely used in lidar in the future, such as integrating photodetectors, front-end circuits, waveform digitization, and waveform SoC for algorithm processing, laser pulse control and other functional modules.

➢ Analog chip: mainly high-precision ADC, which is needed in key subsystems such as lighting control, photoelectric signal conversion, and real-time processing of electrical signals. Mainstream overseas suppliers include TI, ADI, etc. Domestic Silicon Power, Shengbang Micro, Chipsea Technology, etc. can all supply related chips, but there are some performance gaps with overseas leading products.

➢Optical components: including MEMS galvanometers, various optical lenses, OPA silicon photonic chips, etc. MEMS galvanometers are mainly supplied by overseas manufacturers, such as HAMAMTSU, Infineon, ST, etc. Domestic Bofur Optoelectronics, Zhiwei Sensing, Yingtang Intelligent Control, Leishen Intelligence, etc. have related product layouts or have been applied. In terms of optical lenses, major domestic manufacturers have mature technologies. The domestic supply chain has reached the international leading level in optical components and has competitive advantages in cost. It can basically replace foreign supply chains and meet the needs of product processing.

There is ample competition among midstream companies, and domestic manufacturers have outstanding strength. The middle reaches of the industrial chain are mainly lidar integrated product manufacturers, with many participants and fierce competition. Lidar is still in the early stages of mass production. With the development of technology paths, performance and cost advantages may be transferred to various manufacturers. The overall competitive landscape has not yet taken shape, which is also the entry point for domestic manufacturers to develop better. Currently, manufacturers with strong competitiveness in the lidar market are mainly concentrated in China, the United States and Europe. Domestically, companies such as Sagitar Juchuang, Hesai Technology, Leishen Intelligence, and Huawei are developing rapidly, and their related products have a relatively high market share. According to Yole data, Sagitar Juchuang's share of the global lidar market has reached 10% in 2021. %, surpassing Luminar and ranking second.

固态补盲——激光雷达新赛道。 2022年5月，亮道智能推出首款纯固态侧向补盲激光雷达LDSatellite™，面向前装量产，拥有 120°×75°超大视场角。上海车展还展出了前向+侧向补盲激光雷达的集成 DEMO实车。随后，禾赛科技、速腾聚创、一径科技、图达通纷纷推出补盲激光雷达，上市时间均定在 2023年内。

FMCW 激光雷达

相比传统的基于飞行时间（ToF）的脉冲激光雷达，FMCW 激光雷达基于相干探测方法，通过发出连续的激光束，并控制激光束的频率随着时间变化，形成一定的调频曲线，可以响应目标运动引起的多普勒频移，提供目标的瞬时视向速度，FMCW激光雷达具有更高的精度和分辨率，同时也具有更强的抗干扰能力。

FMCW激光雷达的接收和扫描中也需要用到硅光芯片，由于硅光芯片既能导电又能导光，因此它既具有光电探测器的功能，又具有调制解调的功能，从而使得整个系统集成化、小型化。 由中国信息通信研究院技术与标准研究所编撰的《车载激光雷达技术与应用研究报告(2023年)》统计，截止到 2023年第三季度，已有36家中国车企宣布使用激光雷达，预计国内将有高达 106款搭载激光雷达的车型上市，占全球同期预计发布搭载激光雷达新车型总数量近 90%，中国车企在激光雷达领域将持续引领着行业的创新。

芯片国产化，激光雷达规模化可期。从系统构成看，激光雷达芯片主要包括发射端芯片、接收端芯片、信息处理芯片等。 经过多年发展，国内发射端 VCSEL芯片已进入成熟量产阶段，但接收端 SPAD 芯片主要被索尼、安森美等企业垄断。因此，国内 SPAD芯片企业，包括阜时科技、识光芯科、芯视界、灵明光子等，加速技术发展，已实现新的突破。2023 年 8 月，阜时科技发布全固态激光雷达面阵SPAD芯片FL6031，采用Stacked-BSI工艺，集成 360x150的SPAD 像素阵列，有效像素超过50k，并满足上车要求。基于 FL6031芯片，苏州光之矩与武汉万集已完成全固态激光雷达的 Demo开发。

信息处理芯片方面，以 FPGA为主，主要实现时序控制、波形算法处理、其他功能模块控制等。 主要厂商为 Xilinx 和 Altera。速腾聚创、图达通、大疆等车载激光雷达产品的FPGA芯片均来自Xilinx。2023年9月，蔚来发布自研的激光雷达主控芯片“杨戬”（芯片型号为 NX6031）。芯片配备 8个9bit的AD 模数采样，每通道采样频率高达 1GHz，能够高效捕获激光回波信号。

3.1 远期展望百亿美金市场，伴随自动驾驶快速成长

激光雷达是自动驾驶系统中一项关键技术，近年全球智能网联汽车产业进入加速发展新阶段，推动着激光雷达市场规模高速增长。 2022年全球激光雷达解决方案市场规模为 120 亿元 ，根据中商产业研究院预测，2023 年全球激光雷达解决方案市场规模将达到 227 亿元，2024 增至 512 亿元。国内方面，随着智能化技术的持续突破和升级，受无人驾驶车队规模扩张、高级辅助驾驶中激光雷达应用渗透率提升、以及机器人及智慧城市建设等领域需求的推动，国内激光雷达市场驶入快车道，2022年市场规模约26.4亿元，预计2026年可增至139.6亿元。

应用端智能驾驶迅速扩容，预计 24 年赶超机器人领域。 据灼识咨询，在激光雷达解决方案的诸多应用场景中，智能驾驶汽车有望从 2022年的34亿元增至2030年破万亿人民币，CAGR高达103%，随之同期市场份额预计从28.7%提升至79.8%，超过机器人领域（22年占比68.2%）占据最大份额。我们认为，随着汽车自动化水平的提升，单车激光雷达搭载数量将不断增加，L3、L4 和 L5 级别自动驾驶或分别需要平均搭载 1颗、2-3颗和4-6颗激光雷达，市场空间较大。

无人驾驶兴起，市场需求与日俱增。近年来，无人驾驶行业发展迅速，激光雷达作为无人驾驶的重要传感器规模增长较快，其在无人驾驶运用原理是通过 3D 建模进行环境感知，然后通过雷射扫描得到汽车周围环境的 3D 模型，运用相关算法比对相邻帧数环境的变化，从而探测出周围的车辆和行人，做出相应的驾驶判断。据 Yole 相关统计，2021 年，全球激光雷达无人驾驶领域的出货量约 24 万个，同比增长71.43%，预计2025年其出货量将达130万个，CAGR达56.2%。

激光雷达被作为智能驾驶的重要卖点，逐渐布局上车。从 2021 年以来，小鹏、蔚来、理想、极狐、智己等国内车企开始逐步推出搭载了激光雷达的车型，单车搭载量在 1-4 颗不等，越来越多的厂商也在计划布局上车，据 Yole 统计，截止2023年第三季度，已有 36家中国车企宣布使用激光雷达，预计国内将有高达 106款搭载激光雷达的车型上市，占全球同期预计发布搭载激光雷达新车型总数量近90%，国内激光雷达车企的数量和规模将在未来一段时间长期领跑。

智驾汽车渗透率稳定提升，L2向L3+过度正当时。 近年来，全球智能驾驶汽车渗透率逐步提升，从 2020 年的 23.4%增至 2022 年 42.7%，同时带动智能驾驶汽车出货量的增加，2022 年全球智能驾驶汽车的出货量达 3440 万台，根据中商产业研究院的预测，2023 年出货量预计可达 4210 万台，渗透率增至51.0%，2025年出货量攀升至5430万台，渗透率达63.5%。国内来看，我国量产乘用车自动驾驶等级正在由 L2 向 L3+过度，2022 年我国在售新车 L2 和 L3 渗透率分别为 35%和9%，预计 2023 年可达 51%和 20%。部分科技公司正逐步研发 L4 级自动驾驶，并在部分城市路段和特定场景下进行测试，但目前高级别自动驾驶仍存在政策法规、安全性、技术成熟度等众多挑战，根据慧博资讯数据，2022 年我国 L4 渗透率为2%，预计2023 年将达到11%。

Fuel vehicles lack power feedback and range, and electric vehicles are more suitable for autonomous driving. Whether fuel vehicles are suitable for using intelligent driving technology can be considered from the following aspects: 1) The power feedback of fuel vehicles is slow, and its power system is a very complex mechanical structure. The speed-up feeling is relatively slow, and the intelligent driving system itself is superimposed to judge the response time. The risk factor will increase in emergency situations; 2) The lead-acid battery in the oil truck only provides power to the engine when it is running. It has a short battery life in the parking state and cannot properly supply the power required by the automatic driving system. , barely supports L2 and below intelligent driving systems, which is not conducive to iterative development; 3) The LIN and CAN bus networks of traditional fuel vehicles can no longer cope with autonomous driving, and need to be upgraded to faster MOST and vehicle Ethernet buses. However, due to the The reuse of vehicle platforms and modules involves many things, and it is difficult to start over from scratch in terms of architecture.

Advanced assisted driving industry: Downstream companies in this industry mainly include OEMs, Tier 1 companies and new power car manufacturers around the world. Lidar is used in mass production vehicle projects, which usually requires the lidar company to reach long-term cooperation with the car manufacturer or Tier 1 company. Generally, the project cycle is longer.

Autonomous driving industry: Foreign autonomous driving technology research started earlier, and in terms of fleet size, technical level and implementation speed, it still has certain leading advantages compared to domestic ones. Domestic research on autonomous driving technology is developing rapidly, with application pilots and projects being implemented continuously, and the gap with foreign companies is constantly narrowing.

Robot industry: The domestic express delivery and instant delivery industry has a larger market capacity than foreign markets. The domestic technological development level of service robots is equivalent to that of foreign countries. In terms of the richness of robot types and the diversity of landing scenarios, domestic enterprises have more advantages.

Internet of Vehicles Industry: Downstream companies in this industry are mainly providers of Internet of Vehicles solutions. Through these companies, Internet of Vehicles services, including lidar, are integrated and sold to local governments and science and technology parks. There are also cases where lidar companies, governments and science and technology parks are directly connected. Thanks to the vigorous promotion of national policies such as "new infrastructure", the domestic field of Internet of Vehicles has developed more rapidly than abroad.

Innoviz: In-depth cooperation with BMW, winning orders of US$4 billion from Volkswagen

Innoviz, an Israeli lidar sensor developer, was founded in 2016 by four co-founders from the elite technical department of the Israel Defense Forces intelligence force. The company focuses on the research and development of solid-state lidar (LiDAR) sensors and is committed to providing innovative solutions for the autonomous driving and advanced driver assistance systems (ADAS) markets.

Innoviz's lidar sensors are highly competitive in the market and have been selected by many well-known automobile manufacturers and autonomous driving technology companies. In 2017, it announced its cooperation with Jabil, the world's third largest foundry, and in 2018 it announced its cooperation with Magna on the BMW project, and will provide lidar for BMW's L3 production cars to be launched in 2021. In August 2022, Israeli lidar company Innoviz announced that it would provide lidar hardware and supporting software to CARIAD, a subsidiary of the Volkswagen Group. The total transaction price reached US$4 billion, and delivery is expected to begin in 2025.

Innoviz has three major product lines: 1) InnovizOne: the longest detection distance is 250 meters, suitable for L3-L5 cars; 2) InnovizTwo: the cost is reduced by more than 70% compared to InnovizOne, while the performance is significantly improved, mainly targeting the L2+ market , L3-L4 is also applicable, and will be used in Volkswagen and other brands; 3) Innoviz360: Performance, cost and durability have been significantly improved, with a maximum detection range of 300m, mainly for L4/5 level autonomous driving, and mass production is planned for 2024.

Luminar: Accelerating global factory layout

Luminar is an American lidar company founded by Austin Russell in 2012, focusing on the field of MEMS hybrid solid-state lidar. The company continues to focus on the 1550nm wavelength, and its main products are Iris and Hydra. Among them, Hydra is the previous generation lidar product, mainly used for testing and development; the company's latest car-grade product Iris will be mass-produced in 2022, and has reached a supply agreement with Volvo for the autonomous driving system to be launched in 2022. Luminar already works with several automakers, including Mercedes-Benz, Volvo, Audi, Toyota Research Institute, Intel's Mobileye, Airbus and key customers such as SAIC and AV operator Pony.ai.

Luminar unveiled a new version of its Iris sensor at an investor day event in March, which the company said has a 300-meter (984-foot) detection range. Luminar is working with Mercedes to integrate Iris Plus lidar into its vehicles. , mass production is expected to begin in 2025. Luminar also plans to open a manufacturing facility in Asia to support production. A factory in Mexico also came online in the second quarter of this year, and Luminar is also expanding another factory in Thailand with contract manufacturer Fabrinet to produce optical components.

Velodyne & Ouster: merged operations, rapid reshuffling of the competitive landscape

American Velodyne is a technology company located in Silicon Valley, California. It provides autonomous driving solutions with lidar sensor technology as its core. The company was founded by David Hall in 1983. It started in the audio business in the early days, and later gradually expanded its business to the laser radar field. Throughout 2021, Velodyne shipped more than 15,000 units, of which more than 2,400 were solid-state sensors, most of which were mechanical lidars. According to the merger announcement on Velodyne's official website in 2022, Ouster and Velodyne completed a "merger of equals" and the new company will continue to use the Ouster name. The merged company will have more than 850 existing customers and a rich patent portfolio. However, with the rapid rise of manufacturers such as Tudatong, Valeo, Hesai Technology, Sagitar, and Huawei, Velodyne's market share, which once occupied more than 80% of the global lidar market, has shrunk.

Aeva: FMCW technology leader, the product is expected to be put into production next year

Aeva's lidar sensors use advanced frequency-modulated continuous wave (FMCW) technology to enable high-precision distance and speed measurements, as well as high-resolution point cloud imaging. This enables the sensor to accurately detect and identify all objects around the vehicle, including pedestrians, vehicles and obstacles, and can detect objects up to 500 meters away. In January 2021, Avea and Japan Denso announced a cooperation agreement. The two parties will jointly promote the mass production of FMCW (frequency modulated continuous wave) lidar with the goal of meeting the needs of the mass market. Aeva's co-founder and CEO said that the company has received approval from the world's top ten car companies, and mass production models are expected to be put into production in 2025.

Space is limited and will be continued. . .

• 
  

• ↓↓↓
• 
  

  Click "Read the original text" to pre-register now!