LiDAR — light detection and ranging — is one of the most important sensing technologies for autonomous systems. By measuring the time-of-flight of laser pulses, LiDAR systems can build precise three-dimensional maps of an environment at hundreds of meters range and with centimeter accuracy. This capability is critical for autonomous vehicles, service robots, drones, and industrial inspection systems that need to navigate and interact with complex, unstructured environments.
The Problem With Mechanical LiDAR
The early generation of automotive LiDAR systems, exemplified by the iconic Velodyne spinning lidars mounted on self-driving car prototypes, used mechanical rotation to sweep a laser beam across the scene. While effective, these systems have fundamental limitations that prevent mass-market deployment. They are expensive — early units cost $75,000 or more — because precision mechanical assemblies require extensive manual calibration. They are fragile, with moving parts subject to vibration, temperature cycling, and eventual wear. And they are bulky, impossible to integrate unobtrusively into a vehicle's body panels.
The automotive industry requires LiDAR at consumer-electronics cost points, measured in tens of dollars rather than thousands, with automotive-grade reliability over ten-year vehicle lifetimes. Meeting these requirements requires a fundamentally different approach: solid-state LiDAR with no moving parts, fabricated using scalable semiconductor manufacturing processes.
Photonic Beam Steering
The key enabling technology for solid-state LiDAR is photonic beam steering — the ability to direct a laser beam in arbitrary directions without mechanical motion. Two main photonic approaches are competing: optical phased arrays (OPAs) and flash LiDAR.
Optical phased arrays work analogously to phased array radar systems. An array of optical antennas emits light, with electronically controlled phase delays applied to each element. By adjusting the phase pattern across the array, the constructive interference maximum — and therefore the direction of the emitted beam — can be steered rapidly and precisely over a wide angular range. OPAs implemented in photonic integrated circuits can steer beams over fields of view exceeding 120 degrees with angular resolution below 0.1 degrees, at switching speeds measured in nanoseconds.
Flash LiDAR takes a different approach: instead of scanning a narrow beam, it illuminates the entire scene simultaneously with a broad flash of laser light, and uses a 2D array of detectors to capture the returned signal. Flash LiDAR simplifies the optical design by eliminating beam steering entirely, but requires highly sensitive detector arrays — typically single-photon avalanche diode (SPAD) arrays — to achieve acceptable range and signal-to-noise performance.
Solid-state LiDAR enabled by photonic integration is not just an incremental improvement over spinning LiDAR. It represents a paradigm shift that will make high-quality 3D sensing available in consumer products for the first time.
Wove Photonic's Approach
At Wove Photonic, we have developed a large-aperture optical phased array platform based on our silicon nitride photonic integration technology. Our OPA chips integrate thousands of individually addressable optical antennas, phase shifters, and distribution network waveguides on a single chip approximately 10mm x 10mm in size. The chip is driven by a custom mixed-signal CMOS driver IC that applies phase corrections at nanosecond timescales, enabling rapid scene scanning.
Our current prototype achieves 120-degree horizontal field of view with 0.1-degree angular resolution at a range of 200 meters for targets with 10% reflectivity. Power consumption is under 5W for the photonic chip and driver electronics combined, making it suitable for battery-powered applications. We are developing a compact module form factor that can be integrated into vehicle body panels, drone fuselages, and robotic end-effectors.
Applications Beyond Automotive
While automotive has captured most of the press attention around LiDAR, the photonic sensing opportunity extends far beyond vehicles. Industrial robots increasingly need high-resolution 3D perception to manipulate objects and navigate dynamic factory environments. Agricultural drones use LiDAR for crop mapping and precision spraying. Infrastructure inspection systems use LiDAR to detect structural defects in bridges, pipelines, and buildings with millimeter precision. Building automation systems use LiDAR for occupancy detection and security applications.
These industrial and commercial applications often have more achievable cost targets and less demanding certification requirements than automotive, making them attractive early markets for photonic LiDAR technology. The performance requirements are also diverse — different applications require different combinations of range, resolution, field of view, and update rate — creating opportunities for specialized products rather than a single universal solution.
The Commercialization Path
The path to high-volume automotive LiDAR deployment runs through several intermediate waypoints: first demonstrations in controlled environments, then validation in low-speed commercial vehicles (airport shuttles, warehouse robots), then qualification for highway autonomous driving applications. Each step requires accumulated field data and iterative design improvements. Companies that can establish early customer relationships and build reliability data across a range of deployment scenarios will have significant advantages in the eventual automotive ramp.
At Wove Photonic, we are pursuing a parallel commercialization strategy: delivering LiDAR modules to early industrial and robotics customers to build reliability data and generate revenue, while simultaneously developing the next-generation automotive-grade platform that will meet the cost and reliability requirements for vehicle integration at scale. This approach allows us to generate real-world performance feedback that directly improves our technology development roadmap.