LiDAR sensors sit at the intersection of three tightly-coupled supply chains: photonic components (laser diodes, photodetectors, MEMS mirrors), electronics (FPGA/SoC, power, IMU fusion), and mechanical/optical housings, all of which converge on integrators serving autonomous mobility, robotics, UAV mapping, and industrial survey [S2].
The Wind Product Chain Database documents 5,154 industries with over 160,000 upstream and downstream relationships, and that same architecture applies cleanly to LiDAR: raw emitters and detectors feed sub-assemblies, which feed module makers, which feed vehicle, robot, and drone OEMs [S3].
Upstream Component Stack: Lasers, Detectors, Scanners, and Optics
905 nm GaAs laser diodes and 1550 nm InP pulsed lasers are the two dominant emitter classes specified for automotive- and survey-grade LiDAR, with 905 nm dominating cost-sensitive units and 1550 nm enabling higher eye-safe pulse energy for long-range automotive and defense use [S2][S4]. Avalanche photodiodes (APDs) and single-photon avalanche diodes (SPADs) form the receive-side detector stack, paired with narrow-bandpass optical filters to suppress solar background.
Beam steering splits into three families: mechanical rotating assemblies (as in the Xaxxon OpenLIDAR, which uses a bipolar stepper ROB-10551 plus 6-wire slip ring spinning the Garmin LIDAR-Lite v3 at up to 250 RPM) [S1], MEMS-mirror solid-state scanners (Livox Mid-360S class), and flash/electronic-steering arrays. The Xaxxon design quotes a 40 m maximum range, 750 Hz sample rate, +/- 2.5 cm accuracy, and 1 cm distance resolution using the LIDAR-Lite v3 module, illustrating the spec envelope achievable with a USD 269.99 open-hardware rotating platform [S1].
Supporting electronics — FPGA or SoC time-to-digital converters, GNSS/IMU fusion modules, and PoE or USB-2.0 interfaces — round out the upstream BOM. The Xaxxon PCB draws 300 mA at 5 V (1.5 W) from the host USB and exposes a simple serial protocol alongside full ROS drivers, which is a useful lower-bound reference for power budget on hobby-grade units [S1].
Mid-Stream Integrators and Module Makers
Module integrators combine the upstream stack into shippable units: Hesai, RoboSense, Livox, Innovusion, Valeo, Luminar, and overseas players such as Velodyne and Ouster. Livox's public catalog lists the Mid-360S and the HAP automotive-grade unit supplied to Xpeng's P5 platform, with open-source LIO-Livox inertial-odometry algorithms published alongside [S4].
Industrial survey and mobile-mapping integrators (Leica, Riegl, Trimble, Teledyne Optech) target different sub-chains: high-power 1550 nm lasers, 1 MHz pulse rates, and airborne/terrestrial scan heads with calibrated GNSS/IMU. The 2026 China LiDAR company ranking singles out Huawei, leveraging its ICT and chip IP portfolio, as a notable entrant pushing solid-state automotive LiDAR [S6].
Criteria-based comparison of the three dominant scanner architectures for integrators sourcing from this chain:
Mechanical spinning (e.g. Xaxxon-style): lowest unit cost (~USD 270 hobby grade, USD 1,000-4,000 industrial), proven robustness, 360° coverage, but bulkier form factor and moving parts [S1]. MEMS solid-state: compact, automotive-rugged, mid-range cost, narrower field of view (often 120° horizontal), and the dominant architecture for series-production ADAS [S4]. Flash/electronic: no moving parts, simplest reliability story, limited range (typically <50 m) and high peak optical power per pulse [S2].
Downstream End Uses: Automotive, Robotics, UAV, Industrial Metrology
Automotive ADAS and robotaxi programs consume the highest volumes and the most stringent spec sets: 200 m range, 0.1° angular resolution, 1550 nm eye-safety class, and ASIL-B/D functional safety. Livox's HAP for Xpeng P5 is a documented instance of an automotive-grade module reaching series production [S4].
Autonomous mobile robots and SLAM — the use case the Xaxxon OpenLIDAR was designed for — run on lower-cost 2D laser scanners. The 715 Hz sample rate at 180 RPM shown in Xaxxon's ROS Gmapping demo illustrates the data rate needed for indoor mapping with the standard 1 m grid resolution [S1].
UAV LiDAR mapping (Sweep-style 3D-printed scanners, DJI L1/L2 payloads, Phoenix Ranger) sits between the two: payload weight under 1 kg, survey accuracy 2-5 cm at 100 m AGL, paired with PPK/RTK GNSS for direct georeferencing. Industrial metrology and security — fence-perimeter scanning, port crane anti-collision, bin-level measurement — round out the downstream tier, often reusing the same module families as robotics [S1][S2].
Sourcing, Standards, and Failure Modes
Specifying LiDAR for a production line means checking three things beyond range: laser class (IEC 60825-1 Class 1 for eye-safe automotive), functional safety rating (ISO 26262 ASIL-B or D for ADAS), and ingress/environmental rating (IP6K7 or IP6K9K for under-chassis automotive mounting) [S2]. Survey-grade units additionally carry IEC 60825-1 Class 3B or higher because operators are trained.
Common failure modes to spec against: window contamination (condensation, dust, bug strike — mitigated by heater elements and hydrophobic coatings), interference between adjacent units on the same vehicle (cross-talk, addressed by time-synchronization and pseudo-random pulse encoding), and photodetector saturation under direct sunlight (handled by narrow bandpass filters plus adaptive thresholding on the TDC) [S2].
For sourcing teams, the encyclopedia displacement sensor reference and inductive sensor background clarify how LiDAR differs from contact and proximity-based measurement, while the broader machine vision supply context tracks the 2026 component-shortage risk that has hit optical and FPGA lines alongside the LiDAR chain.
Selection Criteria and Who LiDAR Is For — and Is Not
LiDAR is the right tool when non-contact distance measurement is needed at >10 m, when 360° or wide-FOV coverage beats a single point, and when the application can absorb the BOM and power cost of a laser scanner. It is the wrong tool for sub-millimetre static gauging (use a displacement sensor or LVDT), for harsh EMI environments with no optical path, or for cost points below roughly USD 50 per axis where ultrasonic or inductive sensing wins. [S1]
Selection matrix for integrators: budget under USD 500 and 2D indoor mapping acceptable → rotating time-of-flight module (Xaxxon-class); USD 500-2,000 with mid-range 3D and ROS integration → Livox Mid-360S class [S4]; USD 2,000-10,000 with 200 m range and automotive-grade certification → Hesai AT128 / Innovusion Falcon / Luminar Iris class; >USD 10,000 with survey accuracy → Riegl / Leica airborne or terrestrial heads.
Supply-Chain Risks and Trackable Signals for 2026
Three signals worth watching through the rest of 2026: (1) 1550 nm pulsed laser diode capacity expansion at Lumentum, Coherent, and II-VI/Coherent-merged lines, which directly gates long-range automotive LiDAR volume; (2) MEMS mirror wafer supply from STMicroelectronics, Infineon, and Bosch — a single-point bottleneck for solid-state units; (3) FPGA supply, where Xilinx (now AMD) and Intel/Altera lead times have governed LiDAR TDC board availability for two consecutive years. [S1]
Trackable next nodes: OEM automotive LiDAR design-win announcements at IAA Mobility 2026 (September), Hesai and RoboSense quarterly shipment guidance, and any Livox HAP second-platform win beyond Xpeng [S4][S6]. The Wind PDB chain map, which already tracks 160,000+ upstream and downstream company relationships, is the canonical reference for mapping any new node back to its supplier base [S3].
The underlying component specifications are covered under capacitive sensor.