What is LiDAR?

LiDAR, or Light Detection and Ranging, is a remote sensing technology that uses light pulses to map an environment. While you can find LiDAR in home security systems, bar code scanners, and facial recognition systems, LiDAR may be best known for its role in advancing fully autonomous driving. Unlike its RADAR and SONAR cousins, LiDAR provides high-resolution 3D data, making it an important tool across industries including automotive, geology, and agriculture. 

In the early 20^th century, RADAR (Radio Detection and Ranging) and SONAR (Sound Navigation and Ranging) were used to measure distance—the first using electromagnetic radio waves and the latter using sound waves. Ping. Then with the invention of the laser in the 1960s, a new distance sensing system was created that harnessed the power of photonics. LiDAR makes it possible to measure, map and interpret our world at the speed of light.  

Made up of three main parts, all LiDAR systems include an emitter to send out the light waves, a receiver to capture the reflected light waves, and a processor to interpret the data. In a split second, thousands of pulsating light waves hit an object, bounce back, and provide precise timing data to interpret exactly what is in an environment and what it’s doing.

By using multiple laser emitters and pulsing light rapidly (hundreds of thousands of times per second), LiDAR systems are able to capture measurements from different angles across a wide field of view. The result is 3D maps that provide precise information about location, distance and movement. 

The idea of blasting lasers through the air may seem very sci-fi, but LiDAR is a common technology currently used in a variety of industries and applications, including:

Autonomous Driving

Working alongside cameras and radar sensors, LiDAR provides essential data for the safe navigation of self-driving vehicles. By creating 3D maps of the area surrounding the vehicle, LiDAR collects precise information about the shape, size and location of objects, making it possible to detect road signs, other cars, obstacles, and pedestrians. LiDAR’s accuracy at a distance also enables autonomous vehicles to track lanes and helps prevent collisions.

Computer Vision

To derive meaning from images and videos, computers need to more details than cameras alone can provide. Data provided by LiDAR gives computers the ability to recognize and interpret the visual world, by fusing depth information with images. LiDAR-assisted computer vision is commonly found in augmented reality, gesture recognition (gaming), and medical monitoring.

Facial Recognition

In systems that require facial recognition for authentication, LiDAR enhances accuracy and reduces spoofing. LiDAR’s highly detailed point clouds contain precise data about the contours and dimensions within facial structure, helping systems distinguish between real faces and 2D images. And since LiDAR uses its own light source, it can map faces in low-light conditions, even darkness.

Robotics

Similar to autonomous vehicles, robots rely on LiDAR to help them navigate their surroundings safely. LiDAR’s detailed 3D maps give robots the precise data necessary to make informed, split-second decisions about their routes through busy and changing environments, such as warehouses and factories. It’s also helpful for enhancing the object recognition and manipulation capabilities of robots working in complex situations such as manufacturing, agriculture, and search and rescue.

Many of us first heard about LiDAR in relation to autonomous driving. Here, level 5 autonomy (human-like performance in every scenario) is the ultimate goal. To achieve it will require the concerted effort of a suite of sensor types, including camera, radar, ultrasonic, and LiDAR. But it’s LiDAR’s accuracy plus range—seeing objects clearly more than 200 meters away, for example—that makes it widely considered as critical for collision and obstacle avoidance.  

There are several types of LiDAR that can be used just within autonomous driving, including Scanning, Rotating, and Flash. In addition, while most current LiDAR systems utilize Time-of-Flight (ToF) detection technologies, many next-generation systems are looking to leverage Frequency-Modulated Continuous Wave (FMCW) detection.

Manufacturers may use a single type, or all of them at once. Go beyond vehicles and the choice of LiDAR types continues to multiply, with a variety of options that support specific needs in range, mobility and environment. 

Below are the definitions of some common terms used for describing LiDAR systems:

• Rotating LiDAR: This involves rotating the light source to capture images from the full scene. The rotation is mechanically driven (i.e. the spinning tops often seen on top of an autonomous vehicle).
• Scanning LiDAR: This involves using advanced technologies (such as phased arrays, diffractive optics, and metasurfaces) rather than mechanical rotation to capture images from the full scene. This is often also referred to as “solid state” LiDAR.
• Flash LiDAR: This involves using a light source array to capture images from the full scene in a single “shot” (much like the flash on a camera). While this technology is more robust than rotating LiDAR, it generally works over a shorter range.
• ToF detection: This method of LiDAR detection relies on using the time of flight (i.e. the time between when the light source is emitted to when it is detected by the receiver) to determine the distance to the detected objects.
• FMCW detection: This method of LiDAR detection relies on sending multiple beams at slightly different frequencies into the scene and measuring the interference between them to determine the distance to the object. This method is more robust than ToF detection with relation to corruption of the data from other LiDAR systems that may be operating at the same time (e.g. on other cars) but requires significantly more computational processing.
• Airborne LiDAR: This describes LiDAR systems that operate from airborne vehicles such as planes, helicopters, and drones.
• Terrestrial LiDAR: This describes LiDAR systems that operate on terrestrial vehicles such as trucks.

In applications where speed and accuracy dominate the list of priorities, LiDAR is the technology of choice. But as long as LiDAR remains the most expensive of the sensing methods, it will likely continue to meet resistance in its journey towards universal adoption.   

As demand for speed and accuracy across all technologies continues to build, LiDAR’s integration throughout systems will grow. The future of world-changing industries such healthcare, transportation, and agriculture will be shaped by innovations made possible with LiDAR.

To reach the next level of LiDAR capabilities, engineers and manufacturers employ the predictive insights of simulation. With optical and photonics software like Ansys Speos, Lumerical, Zemax, and AVxcelerate, designers can explore numerous scenarios to accurately foresee how LiDAR will perform. Through simulation, testing and validating move light-powered ideas forward even faster.

LiDAR trends to watch for:

• Smart city planning
• Unmanned aerial vehicles (UAVs)
• Environmental research and conservation
• Wearable integrations

Whether it’s looking for pedestrians crossing a street or sinkholes encroaching on a city, LiDAR sees what’s around us with incredible accuracy and speed. By enlisting the power of LiDAR across technologies, we can look forward to many great developments ahead. 

Related Resources