Solid-State LiDAR Components: $22.31B by 2035, and the Margin Sits in ASICs

The 19.30% CAGR measures component demand across laser sources, photodetectors, beam steering, optics, and processing ICs. It does not measure the survival odds of the vendors assembling them.

The solid-state LiDAR materials and components market, valued by Kaiso Research at $3.82 billion in 2025 and forecast to reach $22.31 billion by 2035 at a 19.30% CAGR, is expanding while its most heavily capitalised pure-play sensor vendors consolidate or fail, which shifts the primary sourcing question from sensor specification to supplier durability and component-layer value capture.

_{MicroVision Bought a Production-Proven LiDAR Business for $33 Million}

Luminar Technologies reached high-volume production for the Volvo Cars EX90 in 2023, then filed for Chapter 11 in December 2025 and sold that same LiDAR business to MicroVision for roughly $33 million, according to a February 3 securities filing. MicroVision was the winning bidder at a Section 363 bankruptcy auction.

Read that sequence again, because the order matters. The technology worked. The sensor shipped in a production vehicle from a premium European OEM. The company still ended in liquidation.

The corporate record is unambiguous about how far that unwinding went. Luminar sold its semiconductor subsidiary to Quantum Computing Inc. on 2 February 2026 and its LiDAR business on 3 February, after which it ceased business operations other than to wind down, with a Plan of Liquidation confirmed on 3 April 2026 and equity holders recovering nothing.

That is the fact procurement teams in this segment now have to price. A component market growing at nearly 20% annually does not protect the firms selling into it, and the automotive qualification cycle is long enough to outlive the supplier that passed it.

_{A $3.82 Billion Market Built From Six Component Lines, Not One Sensor}

The market Kaiso Research sizes at $3.82 billion in 2025 is not a sensor market. It is the materials and components layer beneath the sensor: laser sources, photodetectors, beam steering modules, optical components, processing ICs and ASICs, and packaging with thermal management.

Materials span silicon and silicon photonics, gallium arsenide, indium phosphide, gallium nitride, optical glass, quartz, sapphire, and advanced packaging substrates. Five architectures compete across that stack: MEMS, optical phased array, flash, FMCW, and hybrid solid-state designs. Each distributes cost differently across those six component lines, which is why architecture choice is a sourcing decision rather than an engineering preference.

This distinction is where most coverage of the segment goes wrong. Sensor vendors are intermediaries. The primary data shows durable value concentrating in the component lines that serve them, and those lines answer to different economics entirely.

MEMS leads the technology segment today, and the reason is unglamorous and commercial. MEMS-based beam steering hits the range, field of view, and resolution automotive ADAS requires while staying inside automotive quality and cost parameters. It does that without the large rotating mechanisms that made earlier scanning designs unsuitable for production vehicle fitment.

Applications spread wider than the automotive framing suggests: autonomous vehicles, ADAS, robotics, industrial automation, drones, smart infrastructure, security and surveillance, and logistics and warehousing. End users span automotive OEMs, robotics manufacturers, industrial equipment makers, aerospace and defence operators, and infrastructure companies. Automotive OEMs dominate procurement volume, because every production vehicle programme incorporating LiDAR creates structured demand across all six component lines at once.

That concentration cuts both ways, and vendors consistently underweight the second edge. Automotive OEM programmes deliver the volume and the value density that make this market worth serving, with reliability standards spanning millions of operating hours that no other application imposes. The same concentration means a single cancelled vehicle programme removes a disproportionate share of a supplier's forecast revenue in one decision. Robotics manufacturers are the growing secondary end user, and they buy on different terms entirely.

_{Processing ICs Capture the Value, and That Is Not Where Anyone Was Looking}

The leading value contribution in this market comes from processing ICs and ASICs, not from the laser. Signal processing converts raw time-of-flight or FMCW detection data into a usable real-time 3D point cloud, and the quality of that conversion sets the ceiling on what the finished sensor can actually perceive. A superb laser paired with weak processing yields a mediocre sensor.

Custom application specific integrated circuit designs built for LiDAR deliver better detection performance at lower power than adapted general-purpose processors. STMicroelectronics expanded exactly this capability in March 2025, pairing processing IC development with silicon photonics for module manufacturers who need automotive-grade optical integration in smaller packages. Broadcom, Infineon Technologies, and onsemi compete in the same layer.

Here is the part worth sitting with. STMicroelectronics, Infineon, Broadcom, and onsemi carry LiDAR content on balance sheets that do not depend on LiDAR. Luminar and the other pure-play sensor vendors never had that option.

So the value layer and the durability layer turn out to be the same layer. A component buyer weighting supplier continuity is pushed toward diversified semiconductor suppliers, and a buyer weighting processing performance is pushed in the same direction. Those two pressures usually pull apart. In this market they reinforce each other.

That convergence is the single most useful finding for anyone building a sourcing strategy here. It means the defensive choice and the performance choice are not a trade-off, which is rare enough in component procurement to be worth restructuring a supplier scorecard around.

_{Consolidation Is Reshaping the Vendor Map, Not Shrinking Demand}

The vendor landscape in mid-2026 looks materially different from the one most 2024 analyses describe. MicroVision now holds the Luminar Iris and Halo sensor portfolios along with engineering teams in Japan, Sweden, and Florida, and it had already absorbed Scantinel Photonics before that transaction. Quantum Computing Inc. acquired Luminar Semiconductor separately.

MicroVision's stated rationale is worth quoting precisely, because it names the competitive pressure the rest of the segment is managing. Chief Executive Glen DeVos framed the integration around streamlining operations and managing costs, and told reporters the combined portfolio was meant to help the company compete against low-cost LiDAR sensors from China.

That is the actual axis of competition, and it is not autonomy timelines. Hesai Technology and RoboSense operate from a Chinese manufacturing base with automotive supply chain depth and cost-competitive production. Hesai expanded solid-state capability and manufacturing scale in November 2024 to serve automotive OEMs and robotics operators simultaneously, which is precisely the application-flexible posture that survives a slow automotive cycle.

Innoviz Technologies enhanced its InnovizTwo sensor for Level 3 and Level 4 programmes in July 2024 and supplies European OEM initiatives including BMW. LeddarTech, Aeva Technologies, and Ouster remain in the mix across different architectures and price points. The solid-state LiDAR vendor tier is consolidating around firms that either own low-cost manufacturing or sit inside a larger corporate structure.

On the component side the roster is steadier, which is the point. Sony Semiconductor Solutions and Hamamatsu Photonics anchor Japanese photodetector and optical supply, ams OSRAM and Infineon serve European procurement, and the processing IC tier remains intact. Component suppliers did not experience 2025 and 2026 the way sensor vendors did.

The transaction structure itself signals how capital now views this segment. Two of the three notable 2026 changes of ownership, the Luminar LiDAR business to MicroVision and Luminar Semiconductor to Quantum Computing Inc., ran through a bankruptcy court rather than a negotiated strategic sale, and MicroVision's earlier Scantinel Photonics purchase followed the same consolidator logic. Distressed acquisition, not venture funding, has become the primary route by which LiDAR technology changes hands. Buyers are acquiring proven portfolios at liquidation prices instead of financing new entrants, which tells component vendors that the customer roster ahead of them will be shorter and better capitalised than the one behind them.

None of this consolidation reduced component demand. It redistributed who buys the components. That is a very different problem from a contracting market, and it is the one sourcing teams are actually facing.

_{Why the 19.30% CAGR Rests on Regulation Rather Than Optimism}

Regulatory mandates, not autonomous-driving enthusiasm, underwrite the forecast in this coverage. ADAS is the dominant application precisely because safety regulation converts driver assistance from a premium option into a structural requirement across mainstream production volume.

Euro NCAP's 2026 protocol changes restructured assessment around a Safe Driving category covering occupant monitoring, driver engagement, and vehicle assistance, with the highest ratings requiring continuous eye and head tracking linked to assistance-system sensitivity. Requirements of that kind cascade into sensor content per vehicle, because a rating target set at the platform level becomes a bill-of-materials decision at the programme level.

Three forces compound from there. Automotive OEMs integrate solid-state LiDAR as a core ADAS Level 2+ sensor because three-dimensional environmental mapping at highway speeds exceeds what radar and camera fusion delivers alone. Industrial robotics in warehousing, manufacturing, and logistics adds a parallel procurement category with different cost tolerances and shorter qualification cycles. Drone delivery and inspection demand lightweight compact sensors at volumes automotive-grade components alone cannot serve.

The regulatory driver is the one that holds under pessimism, and that asymmetry deserves more weight than it usually gets. Autonomy timelines have slipped repeatedly across this industry, taking vendor revenue forecasts with them. Euro NCAP protocol cycles and vehicle safety mandates arrive on published schedules regardless of whether any robotaxi programme ships.

_{Millions of Validation Hours, Not Unit Price, Set the Adoption Ceiling}

The barrier holding solid-state LiDAR out of mainstream vehicle platforms is qualification duration rather than component cost, and the two get conflated constantly. Automotive component qualification requires reliability validation across millions of operational hours, which stretches the path from design to production fitment by years regardless of how cheap the bill of materials becomes.

Cost is still a genuine constraint, and it concentrates in specific materials. III-V semiconductor laser sources and high-performance photodetectors carry manufacturing costs that make automotive-grade performance at volume price points difficult to reach simultaneously. Hitting the performance target is achievable. Hitting it at the price a mainstream platform will bear is the harder problem.

There's a third barrier that gets less attention than it deserves, and it is strategic rather than technical. Some automotive OEM programmes continue to argue that camera-only perception stacks will suffice, which creates procurement uncertainty across the segment even for OEMs that have already committed to LiDAR. A component vendor cannot forecast demand confidently while the sensing architecture debate stays unresolved at the platform level.

Stack those three together and the compounding effect becomes clear. A vendor commits capital to a multi-year qualification programme, absorbs III-V cost structures that resist compression, and carries the risk that the customer's architecture thesis changes mid-cycle. That combination is what turned an expanding component market into a lethal environment for undiversified sensor vendors, and it explains why the failures clustered at the sensor tier rather than the component tier.

_{Silicon Photonics Is the Cost Curve That Decides the Next Qualification Round}

Silicon photonics is the fastest-growing material in this market, and the mechanism is manufacturing rather than performance. Fabricating photonic integrated circuits on standard CMOS wafer lines lets module designers escape the cost structure of discrete III-V component assembly, which is what has kept automotive-grade LiDAR expensive.

Research institutions have been demonstrating the integration path for years rather than months. imec's work on chip-level optical beam forming targets putting LiDAR components onto a single integrated circuit to enable compact, mass-producible 3D scanning for vehicles, drones, and industrial systems.

Peer-reviewed work has taken the same approach further into the coherent domain, demonstrating FMCW LiDAR on a silicon photonic chip with solid-state beam steering. FMCW matters commercially for a specific reason: it measures distance and radial velocity simultaneously, and it rejects interference in ways time-of-flight designs cannot.

Consider what that means in traffic rather than in a lab. When many vehicles on the same road run active sensors at once, interference rejection stops being a specification line item and becomes a functional safety requirement. That is the transition pulling FMCW from research into automotive roadmaps.

The friction is real and it is not resolving quickly. Integrating silicon photonics with III-V laser sources inside a compact automotive-grade module demands packaging and interface engineering at the frontier of semiconductor manufacturing, and high-quality gallium arsenide and indium phosphide wafer supply remains constrained relative to projected demand. Thermal management inside compact modules across automotive temperature ranges compounds it further.

_{Asia-Pacific Builds It, North America Invents It, Europe Regulates It}

Regional roles in this market are specialised rather than competitive, and confusing the three leads to bad sourcing assumptions. Asia-Pacific holds the leading production position through manufacturing scale, component supply chain depth, and a large domestic automotive market.

China drives the majority of regional procurement and production, with Hesai and RoboSense operating inside that domestic supply base. Japan contributes established automotive electronics expertise through Sony Semiconductor Solutions and Hamamatsu Photonics. South Korea adds semiconductor and automotive volume.

North America concentrates technology development and autonomous vehicle programme investment, and it is where the 2025 to 2026 consolidation played out. MicroVision, Aeva Technologies, and Ouster are based there, and LeddarTech operates from Canada. Innovation leadership and commercial durability proved to be separate questions.

Europe supplies the regulatory pull and a stable component base. Euro NCAP requirements drive structured ADAS procurement across German, French, and Swedish vehicle programmes, while ams OSRAM and Infineon serve component demand. LAMEA is earlier stage, with UAE and Saudi Arabia smart city programmes creating initial fixed-sensor infrastructure demand.

_{What Sourcing Teams and Component Vendors Should Change Now}

Two audiences face different decisions from the same data. For automotive OEM sourcing teams, the practical change is to weight vendor financial durability and portfolio breadth alongside range and resolution, because a qualified LiDAR sensor cannot be second-sourced on a useful timescale. Dual-sourcing at the component layer, where diversified semiconductor suppliers operate, is more achievable than dual-sourcing at the sensor layer.

Concretely, that means asking questions procurement checklists in this category have historically skipped. Does the supplier's LiDAR revenue depend on a single OEM programme? Does the parent balance sheet carry non-LiDAR business? What happens to inventory, tooling, and engineering talent if a Section 363 auction becomes the exit path?

Luminar's Volvo programme answered none of those in advance, and the answer arrived as a court filing. Across Kaiso Research's coverage of automotive sensing procurement, supplier-continuity diligence remains the least formalised part of an otherwise rigorous qualification process. It is usually the last question asked and the first one that matters when a vendor misses a payment.

For component vendors, the non-automotive channels are strategic rather than opportunistic. Industrial robotics, drones, smart infrastructure, security, and logistics procurement operate outside multi-year automotive qualification cycles, so an application-flexible portfolio converts revenue timing risk into something manageable. Vendors serving only automotive absorb the full qualification cycle before earning anything, which is precisely the exposure that ended the pure-play era.

Mid-market robotics and industrial equipment manufacturers running compact sensing requirements will benefit from automotive-driven cost reduction without paying automotive qualification overhead. That is the quiet arbitrage in this forecast, and it favours buyers who aren't bound to vehicle programme timelines.

One caveat belongs on the record, because the primary data does not settle it. Component-layer dual sourcing reduces continuity risk but doesn't eliminate switching cost, since a processing IC change can still trigger revalidation of the perception stack built on it. The exposure moves from catastrophic to manageable. It does not disappear, and any sourcing team told otherwise is being sold something.

_{Where This Market Lands by 2035}

The forecast in this coverage runs from $3.82 billion in 2025 to $22.31 billion by 2035, and the composition of that growth matters more than its slope. Vehicle autonomy, industrial robotics, and drone applications scale concurrently rather than sequentially, which is why the primary data shows demand holding through a period of visible vendor failure.

Silicon photonics and custom ASIC development will keep compressing module cost across the forecast period, moving LiDAR from premium trim into mainstream platforms. Asia-Pacific keeps the production lead. The component layer keeps the value.

Expect further sensor-vendor consolidation before 2030.

_Conclusion

The instinct when a flagship vendor collapses is to question the technology. That instinct is wrong here, and acting on it would be expensive. Luminar's Iris sensor entered volume production on the Volvo EX90, met automotive quality requirements, and shipped, and the component demand it helped create carried on without the company that created it.

What the 2025 to 2026 consolidation actually exposed is a mispriced risk. This industry spent a decade evaluating LiDAR suppliers on range, resolution, and field of view, and treated corporate survival as a background assumption rather than a scored criterion. A $33 million price tag on a production-qualified LiDAR business ended that assumption permanently.

The component lines beneath the sensor, the processing ICs, the photonic integration, the III-V materials, are where both the value and the durability now sit. Sourcing teams still evaluating this market one sensor specification at a time are answering a question the market stopped asking in February 2026.

---

About Kaiso Research and Consulting

Kaiso Research and Consulting is a global market intelligence firm publishing 5,000+ research reports across 11+ industry verticals.

[email protected] | +1 872 219 0417

Dhwani Sharma, Lead Industry Analyst, Kaiso Research and Consulting | Covering sensors, semiconductors, and automotive sensing markets across North America, Europe, and Asia-Pacific

Published: 2026-07-21 | Report Code: SESC1572

Market Study: Access the full index or request a complimentary sample directly via the Global Solid-State LiDAR Materials & Components Market Size, Opportunity Analysis and Forecast, 2026-2035 page