Designing a LiDAR therefore requires combining optics, photonics, electronics and system-level analysis.
How does LiDAR work?
In the simplest picture, the transmitter emits light and the receiver detects the signal returning from an object. In a time-of-flight system, distance can be related to the time of flight:
R = cΔt / 2
where R is the distance, c is the speed of light, and Δt is the time between emission and reception of the signal. The factor of 2 comes from the path sensor → object → sensor. In a real device, though, this equation is the simplest part of the problem.
The LiDAR transmitter
The transmit path can include: laser → beam-shaping elements → scanner → protective window → environment. The design has to control, among other things:
- beam divergence;
- beam diameter;
- emission direction;
- energy distribution;
- losses in the optical elements;
- the scanned field of view.
Lambda Research points to beam divergence, wavelength and detector properties as important parameters when designing LiDAR systems.
The receiver is often harder than the transmitter
After reflecting off a distant object, only a tiny fraction of the emitted energy returns toward the system. The receiver has to collect it and direct it onto the detector. The path might look like: object → aperture → lenses → filter → detector. You have to analyse light-collection efficiency, field of view, spot size on the detector, alignment tolerances, vignetting, parasitic reflections, and optical background.
TracePro — simulating the complete LiDAR assembly
Lambda Research publishes dedicated materials on designing LiDAR with TracePro. The program can perform non-sequential ray tracing of the complete opto-mechanical assembly. The model can include transmit and receive elements, windows, housings and other mechanical parts. This matters in particular because, in a real LiDAR, a ray does not always follow the "intended" path.
Stray light — light the designer didn't want
One of the key problems in LiDAR is stray light, i.e. light entering the system by an unwanted route. Sources can include reflections from the protective window, housing elements, detector mounts, internal surfaces, sunlight, other vehicles' headlights, and road infrastructure. Lambda Research notes that reflections from optical windows, scanner housings and detector mounts can cause ghost signals and degrade the accuracy of distance measurement. Stray-light analysis is therefore not a cosmetic add-on — it can determine whether the system recognizes a real object or an unwanted signal instead.
Why isn't an idealized lens model enough?
A classical optical design might assume a perfectly aligned path. A real device, however, has a housing, screws, mounting elements, edges, adhesive, protective surfaces, and mechanical tolerances. TracePro's non-sequential model can take these elements into account as well. This matters especially in a compact automotive module, where a small change in geometry can open up an entirely new, unwanted optical path.
OSLO — designing the LiDAR's lenses
TracePro doesn't have to be the first tool used in the design. If the receive path requires an imaging system or a carefully designed optic, the lenses can first be optimized in OSLO, Lambda Research's tool for designing and optimizing optical systems. TracePro can then take over the design and extend the analysis to the complete geometry, reflections and stray light. This is a natural split: OSLO — lens-system quality → TracePro — the real module and stray light.
Scene simulation
A LiDAR design shouldn't stop at the surface of the last lens. It's worth analysing the interaction of light with the surrounding scene: vehicles, pedestrians, road surfaces, signs, infrastructure elements. Lambda Research's materials on LiDAR show modelling the system in the context of a real scene and analysing rays propagating between the sensor and the environment. This lets you check how much light actually returns to the receiver and what additional paths arise within the scene.
The automotive environment
An autonomous system doesn't only have to work in a dark lab. It may operate in full sunlight, at dusk, under street lighting, in the presence of other vehicles, with glare from a wet road, and near other active optical sources. Lambda Research points out directly that ADAS and autonomous-platform sensors have to keep working across a wide range of environmental conditions, and that sunlight, headlights and road-surface reflections can lead to unwanted light reaching the sensor.
OptiSystem — the LiDAR's signal-level model
Another part of the problem is signal analysis. Optiwave has training materials on sensor design in OptiSystem, covering LiDAR applications as well. OptiSystem lets you analyse the path at the block level, i.e. combine models of the source, channel, receiver and signal analysis. This is a different approach from ray tracing. TracePro mainly answers the question: where does light go, and what paths does it create in the real geometry? A system-level model answers: how does the transmitted and received signal behave along the measurement path? The two kinds of analysis can therefore complement each other.
Receiver electronics
After optical detection, the signal has to be amplified and processed. The receiver can include a photodiode, an APD, a transimpedance amplifier, a filter, a comparator, and readout electronics. If you need to analyse the electrical and optical signals together, you can bring OptiSPICE into the design.
Testing the worst case
A good LiDAR design shouldn't be optimized only for an ideal target sitting squarely in front of the vehicle. You also need to check a weakly reflecting object, a target at the edge of the FOV, strong off-axis light, a lens shift, a reflection off the protective glass, a change in a coating's properties, and mounting tolerances. This is exactly where a digital model brings a big benefit — many scenarios can be analysed before building yet another costly prototype.
Optimization
The designer isn't optimizing a single parameter. They may simultaneously need high energy on target, high receive efficiency, low stray light, a specific FOV, a compact module, and limited source power. TracePro includes optimization tools that let you build objective functions based on flux, irradiance, spot size, or spatial distribution, among other metrics.
From model to vehicle
A practical design workflow might look like this:
- OSLO — designing the lenses and imaging path.
- TracePro — the complete opto-mechanical model, transmitter, receiver, scene, stray light.
- OptiSystem — modelling the optical path and signal.
- OptiSPICE — driving and receiving electronics.
- Prototype and measurement — validating the real module.
This illustrates well that LiDAR is not a single optical component, but a multi-domain system.
Summary
Designing a LiDAR for an autonomous vehicle requires far more than sketching out a laser–object–detector path. You have to simultaneously control beam shape and propagation, the transmit and receive optics, the returning signal's energy, the field of view, tolerances, internal reflections, stray light, environmental conditions, and the receiver electronics. TracePro from Lambda Research is especially useful for analysing the complete opto-mechanical system and stray light, while OSLO lets you design the imaging elements. Optiwave's tools can complement this analysis on the side of the photonic signal path and optoelectronics. As a result, the design passes through several levels: lens → optical module → signal path → electronics → complete LiDAR.
Manufacturer materials and illustrations
Lambda Research material with graphics and examples on simulating a complete LiDAR: Lambda Research – Simulating and Optimizing LiDAR Systems with TracePro
A newer manufacturer material focused directly on stray-light analysis in LiDAR systems: Lambda Research – Optical Design and Stray Light Analysis for LiDAR Systems Using TracePro
A Lambda Research webinar on modelling automotive LiDAR systems: Lambda Research – Optical Modeling of Automotive LiDAR Systems
An Optiwave webinar on sensor design in OptiSystem, covering LiDAR: Optiwave – Optical Sensor Design using OptiSystem