2026 LiDAR glossary: 9 emerging terms to know

LiDAR vocabulary mixes physical architectures, software functions and marketing names. Capitalization or the word “digital” does not guarantee a standardized technology. This glossary states what each term describes, what it does not prove and what a buyer should verify.

A SPAD-SoC integrates single-photon avalanche diodes, timing/counting logic and often some processing on one chip. Integration can reduce size, analog interconnects and potential cost. Performance still depends on photon-detection efficiency, dark count, crosstalk, dynamic range and ambient-light rejection algorithms.

“Full-color 6D LiDAR” generally means points carrying 3D position, motion/velocity information and aligned RGB color. The six-dimension count varies by supplier: request the exact definition, velocity resolution, colorization method and calibration. It is not a standardized metrology category.

Physical AI covers models that perceive, reason and act in the physical world: robots, vehicles and machines. For LiDAR, it shifts value from raw sensing toward synchronized, semantic, real-time actionable data. The term describes a complete system, not a LiDAR measurement principle.

An Active Camera illuminates the scene rather than relying only on ambient light like passive RGB. It may produce depth, intensity and sometimes color through ToF, structured light or fusion. Check outdoor range, eye safety, interference, exposure, frame rate and multi-sensor behavior.

Smart LiDAR means some filtering, detection, tracking or compression runs inside the sensor. This can reduce bandwidth and latency but may hide raw data and create firmware dependency. Ask which outputs remain accessible, how models are updated and what latency is guaranteed.

Digital LiDAR may describe an architecture where detection and time conversion are tightly integrated digitally, notably with SPADs and on-chip processing. Some vendors use it more broadly to contrast their platform with a conventional analog chain. Compare documented architecture and measured performance, not the label.

Native color LiDAR associates color with depth measurements inside the sensor or a factory-calibrated unit rather than through later external colorization. “Native” does not necessarily mean the same pixel or instant: verify parallax, RGB frame rate, synchronization, night behavior and raw-data access.

Coherent LiDAR mixes the optical return with a local oscillator to extract phase or frequency. In FMCW systems it can directly measure range and radial velocity and reject some incoherent light. Challenges include laser coherence, scanning, photonic integration, cost and signal processing.

Kosmo is a Hesai preview name for a prospective perception platform/category rather than a settled industry specification. Until a final product datasheet is published, treat range, resolution, color or 6D capabilities as claims to confirm. Distinguish demonstrator, engineering sample and qualified product.