China's LiDAR Suppliers: The Sensors Powering Robots & Cars

Searching for "Chinese LiDAR suppliers" usually starts with a car question — but the fastest-growing answer is now about robots. Chinese companies made roughly 95% of the world's automotive LiDAR in 2025, and that same hardware is becoming the default eye for humanoids, robot dogs, lawnmowers and warehouse bots. Here is who makes it.

Why China dominates

LiDAR used to be a niche, Velodyne-dominated market with five-figure price tags. China flipped it into a commodity by controlling three things at once: chip design (SPAD-SoC), manufacturing scale and brutal cost engineering. RoboSense's average selling price fell 52% year-on-year in early 2026 to about US$197 — at that price, LiDAR stops being a luxury and becomes a standard robot component.

The leading players

Hesai (禾赛) — the volume king

Hesai is the global #1 in ADAS main LiDAR, holding roughly 51% of the Chinese market by February 2026 and 13 consecutive months at the top. It shipped more than 40% of the market in 2025 and pioneered the "6D full-color LiDAR" era with its in-house Picasso SPAD-SoC, which fuses point clouds with RGB for what it calls spatial intelligence. Its ATX unit broke into Japan's Toyota ecosystem — a rare foreign win for Chinese sensor makers.

RoboSense (速腾) — the robotics champion

RoboSense (HKEX: 2498) is the most important name in robot LiDAR right now: it ranked #1 globally in robotics LiDAR shipments in Q1 2026, and its robotics business grew 510% year-on-year in H1 2026 while total sales hit 719,200 units (+169.6%). Its E1R all-solid-state LiDAR — compact, cheap and automotive-grade — is inside robot lawnmowers, cleaning bots and humanoids, with a record three-year, 1.2-million-unit order from KUMAR in 2025. The E2 and its EOCENE SPAD-SoC chips (Phoenix, Peacock) push robot perception toward image-grade 3D.

Seyond (图达通) and Huawei

Seyond holds third place (~15.8%) with its long-range Robin series and is the fastest-growing premium player (+385% in Q2 2026). Huawei is the other giant — it pairs LiDAR with its own vehicle compute, holding roughly 24-26% share and shaping the market through its broader intelligent-driving ecosystem. Together Hesai, Huawei, Seyond and RoboSense control well over 90% of the market.

The robot-native players

Beyond the auto giants, a second tier feeds robots directly: Slamtec (思岚) in Shanghai dominates 2D LiDAR for AMRs and vacuums, and YDLIDAR supplies low-cost 2D units from a few dollars up for makers and education robots. For robotics buyers, the split is simple: 2D LiDAR from Slamtec/YDLIDAR for indoor navigation; 3D solid-state from RoboSense/Hesai for perception-heavy machines.

Why this matters for robots

The humanoid and quadruped boom needs cheap, reliable 3D perception. Unitree ships its own 4D LiDAR; RoboSense and Hesai are betting their automotive-scale economics will make "eyes" cheap enough for every robot. China's LiDAR duopoly is exporting its car-supply-chain playbook to robotics — the same story as harmonic drives and dexterous hands: components that used to be exotic are becoming commodities built at Chinese scale.

What to watch

• SPAD-SoC commoditization — when image-grade chips drop below US$100, every humanoid ships with stereo 3D as standard.
• Robot-native form factors — RoboSense's E1R/E2 and Hesai's FTX are designed for robots, not cars; watch which wins the humanoid supply chain.
• The 6D full-color race — fusing point clouds with RGB could make tactile-visual AI much cheaper to train.