Lidar and radar: key differences, operating principle and where they are used in cars

When it comes to modern driver assistance systems (ADAS) or self-driving cars, two terms are more common than others - lidar and radar. Both devices are used to detect objects, measure distances and create a map of the environment, but they operate on completely different physical principles. If you ever wondered why Tesla abandons lidars in favor of cameras and radars, and Waymo or Mobileye continue to actively use them - the answer lies in the technical nuances, which we will discuss below.

In this article we will look at:

• 🔹 Physical principles operation of lidar and radar - how does a laser beam differ from radio waves?
• 🔹 Accuracy and Resolution: Why does the lidar see a pedestrian at 200 meters, but the radar does not?
• 🔹 Weather conditions: How do rain, snow and fog affect each of the sensors?
• 🔹 Cost and widespread use: Why are radars cheaper, but lidars are still a luxury?

If you are planning to upgrade your car with systems ADAS or just want to understand how modern self-driving technologies work - this article will help put everything into perspective. Let's start with the basics.

1. Operating principle: laser vs radio waves

The main difference between lidar and radar is the type of radiation they use to scan their surroundings. This determines not only their capabilities, but also their limitations.

Radar (RAdio Detection And Ranging) works on the basis radio waves — electromagnetic radiation with a wavelength from 1 mm to 1 m (in automotive radars usually 24–77 GHz). The radar antenna emits short pulses that bounce off objects and return back. The signal delay time allows you to calculate the distance to the object, and the Doppler effect (change in the frequency of the reflected signal) allows you to calculate its speed. For example, radar Bosch MRR in Mercedes-Benz can determine the speed of oncoming cars with an accuracy of 0.1 m/s.

Lidar (Light Detection And Ranging) uses laser pulses in the infrared or visible spectrum (usually 905 nm or 1550 nm). The laser beam is reflected from the surface of objects, and the sensor records the time when the light returns. Since the speed of light is known (about 300,000 km/s), lidar can calculate distance with millimeter precision. For example, lidar Velodyne HDL-64E produces up to 1.3 million points per second, creating a three-dimensional map of the environment.

Key difference:

• 📡 Radar works with radio waves, which are weakly absorbed by the atmosphere and can “see” through some obstacles (for example, plastic bumpers).
• 🔦 Lidar uses light that reflects off surfaces with high precision, but is highly scattered by fog or smoke.

⚠️ Attention: Some modern radars (eg 4D radars from Continental) can determine not only distance and speed, but also the height of an object, approaching the functionality of lidars. However, their resolution is still inferior to laser systems.

2. Accuracy and resolution: who sees better?

If we compare lidar and radar in terms of accuracy, then lidar wins by a large margin. The lidar resolution allows you to create 3D point clouds with detail down to several millimeters, while radar provides only approximate contours of objects.

Let's look at the example of pedestrian detection:

• 🚶 Lidar “sees” a person as a set of points, accurately determining the position of the arms, legs and head. This allows the system to recognize whether the pedestrian is walking, running or standing still.
• 📡 Radar detects a pedestrian as a “spot” at a certain speed, but cannot distinguish details. For example, radar will not distinguish a person from a bicycle at the same distance.

Parameter	Lidar	Radar
Distance measurement accuracy	±2–5 mm	±10–50 cm
Angular resolution	0,1°–0,4°	1°–3°
Maximum detection range	up to 250 m (for cars)	up to 200–300 m
Ability to recognize shape	Yes (3D model)	No (outline only)

Lidars are able to distinguish the materials of objects based on the reflectivity of the laser. For example, a metal sign will reflect light differently than asphalt or a pedestrian's clothing - this helps systems better classify objects.

However, radar has its advantage: it works better at high speeds. For example, radar Delphi ESR can reliably track objects at speeds up to 250 km/h, while some lidars begin to "lose" targets when traveling faster than 120 km/h due to scanning limitations.

3. Weather influence: rain, snow, fog

One of the key differences between lidar and radar occurs in adverse weather conditions. This is where radar has a significant advantage.

Lidar strongly depends on the transparency of the air:

- Fog, rain or snow scatter the laser beam, reducing the detection range by 2–5 times.

- Dust or dirt on the sensor can completely block its operation.

- Bright sunlight (especially on snow) interferes with infrared lidars.

Radar practically not subject to these problems:

- Radio waves pass through rain and fog without significant loss.

- Snow can create false alarms, but modern algorithms filter out such interference.

- The radar is able to “see” through light obstacles (for example, bushes or plastic fences).

Why did Tesla abandon lidars?

Elon Musk argues that a combination of radar and neural network cameras can provide the same level of security as lidar, but at a lower cost. However, critics note that without lidar, the system loses accuracy - especially in urban environments with large numbers of pedestrians and cyclists.

⚠️ Attention: In heavy rain or snowfall, even the radar may malfunction. For example, raindrops on the bumper can create “false” targets that the system will mistake for obstacles. Therefore modern ADAS always use a combination of sensors.

4. Cost and mass application

One of the main factors why radars are more widespread than lidars is price. For 2026:

• 💰 Car radar (for example, Bosch LRR4) costs from 50 to 200 dollars depending on the model.
• 💰 Lidar (for example, Velodyne VLS-128) costs from 1,000 to 10,000 dollars.

The high cost of lidars is due to:

• 🔬 The complexity of producing laser diodes and photo detectors.
• 🛠️ The need for precise calibration and assembly (even micron offsets affect accuracy).
• 📉 Low production volumes compared to radars.

However, the situation is gradually changing. Companies like Luminar or Innoviz are developing lidars for the mass market at a cost of about 500 dollars. For example, lidar Luminar Iris already used in Volvo EX90 and Polestar 3.

Study the climatic conditions of the car|Decide your budget (radars are cheaper, lidars are more accurate)|Check compatibility with your car model|Please note that for full autonomy (Level 4–5) lidars are almost mandatory-->

Interesting fact: some manufacturers, for example Toyota, use hybrid systems, where the lidar only works at low speeds (up to 50 km/h), and the radar is turned on on the highway. This allows you to reduce costs without sacrificing security.

5. Application in automobiles: where is what used?

Lidars and radars are used in different systems ADAS, and their roles often complement each other. Let's look at typical scenarios:

System	Sensor used	Example car
Adaptive Cruise Control (ACC)	Radar (77 GHz)	Tesla Model 3, BMW 5 Series
Automatic Emergency Braking (AEB)	Radar + camera (lidar rare)	Volvo XC60, Subaru Outback
Lane Keeping Assist (LKA)	Camera (no lidar needed)	Toyota Camry, Hyundai Tucson
Self-driving (Level 4–5)	Lidar + radar + cameras	Waymo, Cruise (GM)

Lidars are most often found in:

• 🚗 Premium cars (Audi A8, Mercedes S-Class) for night vision or parking systems.
• 🤖 Self-driving taxis (Waymo, Yandex Self-Driving) where maximum accuracy is required.
• 🏗️ Mapping systems (for example, Mobileye uses lidars to create HD maps).

Radars have become standard equipment even in budget cars. For example, Renault Arkana or Kia Rio equipped with radars for systems ACC and AEB already in basic configurations.

6. The future of technology: what awaits lidars and radars?

Experts agree that in the next 5–10 years, lidars will become cheaper and more widespread, but they will not be able to completely replace radars. Here are the key trends:

Lidars:

• 📉 Price reduction up to $200–300 by 2027 (according to forecasts Yole Développement).
• 🔄 Solid state lidars (without moving parts) will replace mechanical ones, which will increase reliability.
• 🌍 AI Integration for better object recognition (e.g. lidars Ouster already use neural networks to filter noise).

Radars:

• 📡 4D radars (with height detection) will become standard in the auto class C and above.
• 🔗 Network radars (for example, NXP S32R) will exchange data between cars to create a collective road map.
• 🚘 Miniaturization: radars will become smaller and cheaper, which will allow them to be installed around the perimeter of the car (as in Ford BlueCruise).

By 2030, more than 30% of new cars are expected to be equipped with lidar, but radar will remain the primary sensor for safety systems due to its reliability and low cost.

FAQ: Frequently asked questions about lidars and radars

❓ Is it possible to install lidar on an old car?

Technically yes, but it would require:

• 🔧 Integrations with ECU (firmware may be required).
• 💻 Software settings for data processing (not all lidars are compatible with older cars).
• 💰 Installation costs (from 1,500 to 5,000 dollars including work).

It is much easier and cheaper to install radar for systems ACC or AEB.

❓ Why doesn't Tesla use lidars?

Tesla bets on visual neural networks (cameras + AI), arguing that:

• 👁️ A person drives a car relying on vision, not laser or radio waves.
• 💰 Lidars are expensive and complicate production.
• 🛠️ Neural networks can learn from large data, improving accuracy over time.

However, critics note that without lidars the system Full Self-Driving (FSD) copes worse with recognizing pedestrians in bad weather conditions.

❓ Which sensor is better for parking?

For parking, the following are most often used:

• 📡 Ultrasonic sensors (cheap, but with a short range - up to 2 m).
• 🔦 Short range lidars (for example, in Audi A8 for 3D parking).
• 📷 360° cameras (they give a visual picture, but do not measure the distance).

Parking radars are rarely used due to their low resolution at close ranges.

❓ Can lidars and radars work together?

Yes, and this is the most effective approach. For example, in Mercedes DRIVE PILOT (Level 3) are used:

• 🔦 Lidar Luminar for high-precision 3D mapping.
• 📡 Radar Bosch to track the speed of objects.
• 📷 Cameras Mobileye for recognizing road signs.

This combination allows you to compensate for the weaknesses of each sensor.

❓ Does the color of the car affect the operation of the lidar?

Yes, but only slightly. Lidars reflect better from:

• ⚪ Light and matte surfaces (white, grey).
• ⚫ Dark glossy surfaces (black metallic).

The worst thing lidars “see” is:

• 🔴 Bright red or orange cars (due to the wavelength of the laser).
• 🌫️ Cars with tinting or dirty windows (laser scatters).

However, modern algorithms correct these distortions.