ZoZo Auto Research has released the "2025-2026 LiDAR (Automotive, General Robotics, etc.) Application Research Report".
The "2025-2026 LiDAR (Automotive, General Robotics, etc.) Application Research Report" mainly covers the following contents: LiDAR overview and development trends, technical structure and components, market data analysis, core component supplier research, leading domestic and international LiDAR providers, LiDAR product parameter summary, LiDAR technology comparison, and LiDAR application research in different fields (automotive, general robotics), etc.
[Introduction]
In 2025, the total installation volume of LiDAR in China's passenger cars reached 3.21 million units, a year-on-year increase of 110.1%.
High Line Count Brings Generational Breakthrough from "Point Cloud" to "Image"
Intensified Competition Makes Combined Perception Solutions Gradually Become the Mainstream
Accelerated Expansion from Automotive to General Robotics Applications
I. In 2025, the Total Installation Volume of LiDAR in China's Passenger Cars Reached 3.21 Million Units, a Year-on-Year Increase of 110.1%.
In 2025, leading automakers represented by BYD, Changan, and Li Auto have continued to advance the "technology democratization" competitive strategy, resulting in advanced intelligent driving functions no longer being exclusive to top-trim models, but being popularized across the entire model lineup. Against this backdrop, as one of the core sensors for realizing advanced functions such as urban NOA (Navigate on Autopilot), the adoption rate of LiDAR has risen accordingly. To support urban NOA, many models have made LiDAR standard across all trims, such as the 2026 Lynk & Co 08 EM-P, Gaoshan 7, AITO M6, and Dongfeng Yijing.
In addition, on January 1, 2028, China will begin to implement the mandatory national standard "Technical Requirements and Test Methods for Automatic Emergency Braking Systems (AEB) for Light Vehicles" with the standard number GB 39901-2025. At that time, all new models must meet the requirements of this standard, while models that have already obtained type approval will have a transition period of 12 to 24 months to complete technical adaptation. The new national standard imposes stricter requirements on AEB performance, especially in complex scenarios such as nighttime. To meet the stricter AEB testing standards in the future, LiDAR has become an indispensable perception hardware, providing policy expectations for the long-term installation rate of LiDAR.
Driven by the widespread adoption of vehicle advanced driver assistance systems (ADAS), by 2025, the cumulative installation volume of LiDAR in China's passenger cars has exceeded 5 million units. Among them, in 2025, the total installation volume of LiDAR in China's passenger cars reached 3.21 million units, a year-on-year increase of 110.1%.
Looking at the monthly data over the past three years, the installation rate of LiDAR in China's passenger cars has maintained a steady upward trend. In May 2025, the LiDAR installation rate broke through 10%, mainly driven by hot-selling newly launched models such as the Li Auto L6 and the 2025 AITO M9. In November 2025, the LiDAR installation rate exceeded 15%, reaching 16.6%, mainly affected by the volume growth of new models such as the Xiaomi YU7 and the 2026 AITO M7. In December 2025, LiDAR hit a new high with an installation rate of 18.5%, with major influencing models including the Fangchengbao Tai 7, 2026 NIO ES8, etc.
II. High Line Count Brings Generational Breakthrough from "Point Cloud" to "Image"
In recent years, the "line count" (vertical resolution) of LiDAR has become a key metric for measuring its performance. Mainstream products have rapidly advanced from 128 lines and 192 lines to higher specifications. For example, ultra-high line count LiDARs include Huawei's 896-line dual-optical-path image-level LiDAR, Hesai ETX (800 lines) and AT1440 LiDAR (1440 lines), as well as RoboSense EM4 which supports customization from 520 to 2160 lines. In 2026, 500 lines has become the "new threshold" for flagship smart models, while 192 lines has trickled down to mass-market models priced under 200,000 RMB, becoming a standard configuration.
"High Line Count" Drives the Industry from "Point Cloud-Level" Perception to "Image-Level" Perception
On March 4, 2026, Huawei Qiankun released a new generation of dual-optical-path image-level LiDAR, pushing the mass-production specification to 896 lines, becoming the highest level globally at present. It is first deployed on the refreshed versions of the Maextro S800 and AITO M9, solving the pain points in long-tail scenarios (Corner Cases) of autonomous driving:
Significantly Extended Detection Range: The 896-line LiDAR increases the perception and detection range for low-reflectivity targets (such as black tires) from 42 meters to 122 meters, and improves the detection range for irregular obstacles (such as traffic cones) by 77%. This provides longer decision-making and reaction time for high-speed driving, fundamentally enhancing active safety capabilities.
Generational Upgrade in Recognition Accuracy: The minimum target height that can be stably recognized is reduced from 30 cm to 14 cm, enabling precise identification of small obstacles on the road that were previously easily overlooked, such as small cardboard boxes, gravel, and fallen traffic cones. In nighttime scenarios, it can even clearly see the details of a dog wagging its tail 55 meters away, achieving an "image-level" perception standard.
High line count directly brings an exponential improvement in angular resolution. At the traditional level, the vertical angular resolution of 192-line LiDAR is usually around 0.2°, while LiDARs with over 500 lines can compress the angular resolution to 0.05° or even 0.01° (e.g., AT1440 0.05° x 0.0125°, AT128 best 0.1° x 0.2°, ATL 0.08° x 0.1°, EMX 0.08° x 0.10° global resolution, EM4 0.050°×0.025°, Falcon K3 up to 0.07°×0.03°).
Among them, Hesai Technology's AT1440 is an automotive-grade ultra-high-definition LiDAR designed for advanced autonomous driving. Its core features include a top-tier 1440-line count and image-level point cloud output capability, with a single-return point frequency of up to 34 million points per second. Equipped with Hesai's fourth-generation self-developed chip, it adopts cutting-edge high-efficiency light sensing and ultra-high parallel processing technologies, achieving a detection range of 300 meters @10%. It is mainly designed for L4 and above autonomous driving systems and is the core sensor of the "Thousand-Millimeter Eye A" perception solution. This solution typically uses four AT1440 units working collaboratively to achieve 360° full coverage and zero-blind-spot perception, with application areas including Robotaxi and Robotruck.
Another example is the Seyond Falcon K3 long-range LiDAR series, which is equivalent to 600 lines, with a maximum angular resolution of 0.07°×0.03° and a detection range of up to 350 meters. It is mainly equipped on NIO's high-end models, providing key perception support for L3 and higher-level autonomous driving, with cumulative deliveries exceeding 600,000 units (as of early January 2026).
Behind the leap in resolution is the paradigm shift from analog architecture to digital chips, and the evolution from analog signal (APD) architecture to digital signal (SPAD-SoC) architecture. SPAD (Single-Photon Avalanche Diode) chips share the same origin and structure with camera CMOS, adopting pixel arrays. This makes increasing the line count just like increasing the pixels in a camera, no longer constrained by the physical limits of the number of analog channels and complex circuits.
Taking RoboSense's EM4 chip as an example, it adopts VCSEL + SPAD-SoC chips, integrating multiple advanced technologies such as digital architecture, crosstalk cancellation, full-condition optoelectronic signal processing, and lossless data compression. Based on platform-based design, EM4 supports customizable technical solutions of 520 lines, 720 lines, 1080 lines, and 2160 lines. Currently, the customized 520-line version, with its mature and mass-producible generational advantage, has achieved mass production and application in multiple models such as IM LS9, IM LS6, and Zeekr 9X.
EM4 can provide vehicles with 1080P high-definition 3D perception capabilities and an imaging capability of 25.92 million points per second. It not only achieves a maximum detection range of 600 meters but also accurately identifies small distant objects such as tires, traffic cones, and cardboard boxes. Compared with current mainstream LiDAR products, EM4 can extend the response time of the intelligent driving system by up to 70%, making the system's decision-making response more relaxed and the intelligent driving experience safer and more comfortable.
The high line count of automotive LiDAR expands the safety boundary of autonomous driving from "high-probability visibility" to "extremely low-probability detail visibility", providing better "safety redundancy". This leap is a key cornerstone for advanced autonomous driving at L3 and above to move from "usable" to "reliable".
III. Intensified Competition Makes Combined Perception Solutions Gradually Become the Mainstream
Currently, global intelligent driving is in a critical period of transition from L2+ to L3. Meanwhile, against the backdrop of overall sales pressure for Chinese automobiles in 2026, LiDAR, as one of the core components of intelligent driving perception, has seen its industry competition upgrade from comparing single LiDAR performance to providing overall solution capabilities of "hardware combination + algorithm synergy + scenario adaptation".
Regarding safety standards for different levels of intelligent driving, taking Hesai as an example, it focuses on ATX products (with a cost of about 200 USD) in the L2 market, promoting the transformation of LiDAR from "high-end configuration" to "safety standard configuration"; in the L3/L4 market, it defines a higher safety ceiling through high-performance combinations, such as the L3 LiDAR combination of ETX*1 + FTX*2, and the L4 LiDAR combination of AT1440*4 + FTX*4.
Other combined perception solutions include:
Seyond "1+2 LiDAR Combination" Solution: Equipped with a combination of 1 Falcon ultra-long-range main-view LiDAR + 2 Robin W wide-angle LiDARs. These three LiDARs are integrated into the body of the new NIO ES8, building a full-scenario perception matrix from long-range to near-field and from main-view to blind-spot filling.
Wanji Technology WLR-760 + WLR-750 Autonomous Driving Perception System: Currently implemented on the Jiushi Intelligent Z5, its perception system consists of 2 front-facing WLR-760 LiDARs and 2 side-rear-facing WLR-750 LiDARs, achieving all-round and high-precision environmental perception. The Z5 is aimed at urban logistics distribution, featuring flexible cargo box configuration and multi-mode distribution capabilities.
MicroVision Three-LiDAR Architecture, 1 MAVIN + 2 MOVIA S: With the Tri-Lidar architecture, MicroVision integrates multiple LiDARs (e.g., two short-range (MOVIA S) and one long-range (MAVIN)) into a unified open-platform design. This design allows the long-range LiDAR (such as the simplified Mavin) to focus on farther detection, while the corner LiDARs are responsible for covering the edges of the field of view, working collaboratively to achieve comprehensive environmental perception.
RoboSense's EM4+E1 combination has become one of the preferred perception solutions for the new generation of Robotaxis, and has completed product mass-production verification with 8 leading global customers. In other fields, for example, on September 15, 2025, RoboSense reached a strategic cooperation with MINIEYE. MINIEYE's new generation of Xiaozhu unmanned vehicles T5 and T8 will be equipped with 3 RoboSense digital LiDARs, including 1 automotive-grade 192-line EMX and 2 full-solid-state blind-spot filling LiDARs E1R, which can significantly improve the perception accuracy and safety redundancy of unmanned vehicles in complex traffic scenarios.
IV. Accelerated Expansion from Automotive to General Robotics Applications
In addition to automotive applications, LiDAR is being deployed on a large scale in various robotic scenarios, providing precise perception capabilities for AGV, AMR, as well as quadruped and humanoid robots. Among them, robotic lawnmowers and unmanned delivery robots have become the main application scenarios. Currently, representative products include Livox MID-360, Lanhai Optoelectronics LDS-M300, RoboSense Airy/E1R, and Hesai JT series, etc.
For example, in 2025, RoboSense's LiDAR product sales in the robotics field reached as high as 303,000 units, widely used in scenarios such as smart robotic lawnmowers, unmanned delivery, and humanoid robots. Compared with 2024, RoboSense's product sales in the robotics field achieved a strong year-on-year growth of 1141.8%.
Taking Hesai's JT series 3D LiDAR as another example, after its release in January 2025, it achieved explosive growth in the robotics market relying on its outstanding performance. It quickly reached the milestone of 100,000 units delivered from its release to May 2025, and achieved cumulative deliveries of over 200,000 units by the end of 2025. The unlocked robotic application scenarios include agriculture (Agtonomy unmanned agricultural machinery), cleaning (Gaussian Robotics), mowing (Dreame), airports (Boenic smart airports), smart room measuring (Rushi), game modeling (Black Myth: Wukong), and factory automatic scheduling (BMW factory in Germany), etc. In the field of logistics robots, Hesai is also assisting Meituan drones, Jiushi unmanned vehicles, Neolix, etc., to bring innovative solutions to urban distribution logistics.
Expanding towards "general robotics", leading LiDAR enterprises can not only reuse their platform-based technologies in the robotics track to support the dual-line growth of "ADAS + robotics", but also diversify business risks and find new growth poles.
[Contact Information] Mobile number is the same as WeChat ID
Industry Research Department | Mr. Fu 15810027571
Mr. Zhao 18702148304
Data Service Department | Ms. Liao 13718845418
Strategic Consulting Department | Ms. Han 15810133447
Promotion and Communication Department | Ms. Zhang 13716037793
Mr. Du 13910162318