Three terms decide almost everything about who can compete where in this industry, and what they are allowed to earn for it. Learn these three and the rest of the value chain organises itself.
Nearly every LiDAR sold today is a time-of-flight (ToF) device: it fires a discrete laser pulse and measures the elapsed time until a reflection returns, using the constant speed of light to convert that time directly into distance. It is conceptually simple, mature, and — because it only measures distance — says nothing about how fast the object it just detected is moving; that has to be inferred frame-to-frame by software. Frequency-Modulated Continuous Wave (FMCW) LiDAR, sometimes marketed as "4D LiDAR," instead emits a continuously frequency-swept beam and measures the beat frequency between the outgoing and returning signal — a coherent-detection technique borrowed from radar — which yields instant, per-point velocity (via the Doppler shift) alongside range, in the same measurement. FMCW's coherent receiver is also structurally immune to interference from another LiDAR's pulses, a property that matters more every year as multiple vehicles' sensors increasingly share the same road. It is also a far harder and more expensive engineering problem — it requires a narrow-linewidth laser and a coherent receiver architecture very few companies have actually shipped at volume.
The single wavelength a company picked years ago is one of the most consequential decisions in this industry, for a reason that has nothing to do with cleverness and everything to do with human anatomy. The human eye's cornea and lens focus 905nm near-infrared light straight onto the retina, so a 905nm laser's maximum safe output power is tightly capped by the IEC 60825-1 Class 1 eye-safety standard. 1550nm light, by contrast, is almost entirely absorbed by the cornea before it ever reaches the retina — the eye is effectively opaque to it — so a 1550nm laser can emit roughly 40 times more power at the same Class 1 eye-safety rating, which translates directly into longer range and better performance in rain, fog and dust. The catch is that 905nm detectors are ordinary silicon — the same semiconductor material every consumer-electronics fab already knows how to make cheaply at volume — while 1550nm detection requires indium gallium arsenide (InGaAs), an exotic, far more expensive III-V compound semiconductor with a much thinner global supply base. Range and cost pull in opposite directions, and a company's wavelength choice is a bet on which one the market will pay for.
A LiDAR that works reliably on a test bench says nothing about whether it is automotive-qualified. That is a separate, much narrower gate defined by ISO 26262, the automotive functional-safety standard, which assigns every safety-relevant vehicle function an Automotive Safety Integrity Level from ASIL-A (lowest) to ASIL-D (highest, reserved for functions like automatic emergency braking where failure can be fatal). Earning an ASIL-B or ASIL-D rating for a LiDAR module means proving, through years of documented engineering and testing, that its failure modes are understood, bounded and — critically — that the system can detect its own degradation before the human or the automated-driving software would otherwise find out too late. It sits alongside, and is reinforced by, IATF 16949, the automotive quality-management certification a component plant needs before any serious OEM will even open a purchase-order conversation. Neither certifies performance in the way a spec sheet's stated range does; both certify that a failure, when it eventually happens, happens in a way the vehicle can handle.
| Term | What it actually certifies | Who typically holds it | What it does NOT tell you |
|---|---|---|---|
| Time-of-flight (ToF) | Distance, via elapsed time of a discrete pulse | The overwhelming majority of LiDAR shipped today, across every company in this report except Aeva | Object velocity — that has to be inferred separately, frame to frame, by perception software |
| 1550nm wavelength | Roughly 40x the safe laser power of 905nm at the same IEC 60825-1 Class 1 eye-safety rating, hence longer range | Companies that bet on long-range highway ADAS/L3 use cases; requires more expensive InGaAs detectors | Whether that extra range and cost actually converts into a design win the OEM will pay a premium for |
| ASIL-B / ASIL-D (ISO 26262) | That a component's failure modes are documented, bounded and self-detecting — not that the component performs well | A smaller set of companies with multi-year automotive engineering programmes behind them; see each company's own report for its disclosed certification status | Whether the certified part has actually won a production contract, or how long that qualification remains valid as the OEM's own platform evolves |
Source: Dart Consultants, from IEC 60825-1:2014 laser product-safety standard documentation, ISO 26262 automotive functional-safety standard summaries, and industry technical literature on 905nm/1550nm eye-safety power limits (Laser Focus World; lslidar.com; inertiallabs.com) — see Notes for full citation list.
If a company's product is a 905nm, time-of-flight, mechanically-scanned sensor selling into short-range industrial or entry-level ADAS use cases, assume it is competing largely on manufacturing cost against a crowded field, much of it Chinese. If a company's product is 1550nm and/or FMCW, aimed at highway-speed automotive or robotaxi-grade redundancy, assume the engineering bar — and the number of companies that have actually cleared it at volume — is far smaller, and that is where this report's attention, and the industry's own consolidation, should be read most closely.
