How lidar sees objects

What is Lidar?

Lidar is surveying or remote sensing method that uses laser light to measure the distance between the target and the sensor. Simply said, it’s like radar but using light instead of radio waves.
Lidar technology in cars

What does LIDAR mean?

Name “LIDAR” has several origins –  acronym of “LIght Detection And Ranging” and a bit older combination of LIght and raDAR. Also, evolution of technology and extended purpose led to development of an additional version – Light Imaging, Detection and Ranging.

If you are wondering – all versions are correct and valid so don’t worry too much about how your write the name.

How does it work? 

Lidar system is designed to do one specific task – to measure the distance between the object and originating point (position of the system).

What is so hard about measuring the distance?

Lidar releases the light in the form of a pulsed laser beams toward the target (object, surface, person) to measure ranges (variable distances) to the target.

Hard part of the whole system is the sheer number of pulses released and points measured. Every second, Lidar system measures the distance for 150.000 pulses released.  

Distance is calculated by measuring the attributes from the reflected light with a sensor and the data is combined into a 3D model (high resolution map) of the target surveyed.

This 3D model of the information received from all the light pulses Lidar released is called a “point-cloud” and each point in it has 3D spatial coordinates (latitude, longitude, height).

Having point cloud as an end result in mind, we can say that Lidar is actually combining 3D scanning with laser scanning, enabling wider usage of the technology.

Lidar point cloud scanning of a bridge

How is distance actually measured with Lidar?

Simple explanation:

Lidar releases the light pulse and measures the time needed for pulse to come back, being reflected from the target. As we know that light moves at a constant speed, distance is the only unknown in this equation:

Distance = (Speed of Light x Time of Flight) / 2

There are 2 LIDAR methods of analysis (detection):

  • time-of-flight (TOF)
  • coherent detection methods

Time-of-Flight principle (ToF) is a method for measuring the distance between a sensor and an object, based on the time difference between the emission of a signal and its return to the sensor, after being reflected by an object.)

https://www.terabee.com/time-of-flight-principle/

By “bombarding” the environment with thousands of pulses Lidar maps them and transforms the data into a digestible form for a human, and this visualisation we call “point cloud”. 

Frequency-modulated continuous-wave (FMCW) is a commonly used coherent detection method. In this method, what is measured is the frequency coming from a local oscillator that beats the received signal (source: “Integrated LIDAR with Optical Phased Arrays in Silicon Photonics” (PDF)).

FMCW method is getting more traction and awareness because it enables measuring the speed together with distance. And additionally, it uses lower cost electronics and simplified build compared to direct ToF measurement systems.

Let’s just summarise how Lidar works on an example when used on a plane:

  1. Pulse is released from the device
  2. Returned signal is collected by sensor in Lidar system
  3. Distance is calculated (Time of Flight calculation)
  4. Plane’s position and altitude are collected by GPS and Inertial Measurement unit and stored
  5. Computation of the precise echo position
  6. Result is integrated into a point cloud (with results from all other pulses released)

If Lidar is used on the ground, data set is enriched with geolocation of the device (starting position) coming through GPS device (module) incorporated into the Lidar system.

When Lidar is used from the air (airborne Lidar) then dataset has to be enriched with changes in height, direction and location.

Who invented Lidar?

Development of Lidar happened due to breakthrough in laser technology in the 1960s. First mention of the system with Lidar fundamental elements happened in 1961, in the aviation industry. System was developed by Hughes Aircraft Company (yes, owned by THAT Howard Hughes) to track satellites. 

Originally called CoLidar (Coherent Light detecting and ranging) it is basically origin for all laser rangefinders, laser altimeters and lidar units developed. 

Military (as it happens with all major breakthroughs) took up the invention and used the terrestrial ranging technology for targeting.

Outside of military purpose, Lidar was first adopted by Meteorologists to measure clouds and pollution. In 1965, Ronald Collins from Stanford Research Institute filed a patent describing a LIDAR (light-radar) system for weather and atmosphere analysis. Globally, lidar was put on the map in 1971 by Apollo 15 expedition in which it was used to map the moon’s surface.

Lidar technology has continued to be developed since then, making the system more compact and smaller in size. In 2015 DARPA a miniature LIDAR system on a single chip, opening a new era in Lidar development.

Are there different types of Lidar?

Yes, and there are several ways how Lidar is classified. First classification structures Lidar systems based on the way how the system is set up to scan the wider target surface.

So, either the system has rotating elements that release the pulses and covers 360′ view around, or it is designed to cover a specific area in front without moving parts.

2 main types of Lidar system based on how they are built:

  1. Rotational Lidar
  2. Solid state Lidar
Lidar systems are used for scanning of Earth's surface
Source: https://oceanservice.noaa.gov/facts/lidar.html

Another classification of Lidar is derived from the way how it is used and for what kind of target area:

  1. Airborne – scanning is done from the air
    1. Topographic – target is Earth surface, manmade or natural environments
    2. Bathymetric – uses green laser that is able to scan below water surface level
  2. Terrestrial
    1. Static – position of Lidar system is fixed during scanning, but usually it has smaller dimensions and it is easily portable
    2. Mobile – most talked about version, due to implementation in self-driving car concepts. System usually combines camera vision, sensors and GPS location tracker
NASA heavy lift helicopter carrying ALHAT lidar equipment, NASA 2010

What is inside Lidar?

Typical Lidar system consists of:

  • Laser
  • Scanner
    • Photodetectors and receivers
  • IMU – inertial navigation measurement unit
  • Navigation and positioning system

How does the laser work in Lidar system?

Lidar scanning starts with the releasing a pulse of light. And exactly this is the role of the laser – to create and emit a light pulse.

How is pulse created by laser?

Quartz flash tube uses high voltage electrical current to burst a light. Light burst starts to move atoms in ruby crystal and when they reach certain energy threshold they start to emit light particles (photons).

Every photon created amplifies creation of other photons from the atom. In the end, photons exit through a silver mirror and together they make a laser light.

What are the types of laser used in Lidar systems?

Depending on the purpose and the environment, laser will be chosen based on its wavelength and Lidar operator has to balance between possible distance and the data quality it wants to receive. 

Most commonly used lasers are operating on 600-1000nm wavelength, due to the possibility of using silicon photodetector which reduces the cost of Lidar system.Their maximum power, on the other hand, has to be reduced when there is a possibility of hitting a human eye due to possible damage to the retina. 

If Lidar is used for the area without risk of people being exposed to it and not in the water, usual frequency used is around 1064nm. Usual combination with this wavelength is using InGaA photodiodes.

In the case when the objective is underwater mapping (bathymetric lidar), 530 – 540nm frequency is used as it penetrates water with less attenuation.

When you read about self-driving car technology and Lidar, this is what you should know about lasers:

Sweet spot for using Lidar in populated areas is higher wavelength (1550nm), as it enables collecting data in the range of 200m, without possible impact on people.

Power and distance trade off  is found to be better compared to laser with 900-1000nm range with lower power consumption. Other elements that impact overall safety are divergence angle, pulse duration, exposure direction.

Due to rotational parts that require large electromechanical components to provide workable results, current development is focused on solid state Lidar. There are several solutions being explored and partially commercialized:

Scanner

Role of the scanner is to measure the angle at which light pulse was fired and to detect reflected pulse from the target.

Most Lidar systems in the past were equipped with rotational scanner. Due to its size and sensitivity to movements, they are not a viable solution for modern applications (self-driving cars).

Size comes from the fact that rotational scanner cannot scan vertically, and instead you need to add additional mirrors to enable it. Adding mirrors, adds complexity and number of components and that leads to a bigger, more complicated and more sensitive system.

At the moment, development on the Lidar system scanners is grouped in 3 topics:

  • Rotational scanner – already mentioned, used for small distances, cheap enough for DIY
  • Optical phase arrays – to have a viable Lidar system with optical phased arrays, it should be supported with high beam power and large aperture size. Optical phased arrays are built on silicon photonics
  • MEMS – the price and resolution of lidar sensors still do not meet the target values for the automotive market to be accepted as a basic sensor for ensuring safe autonomous driving. Recent work has focused on MEMS scanning mirrors as a potential solution for affordable long range lidar system (https://link.springer.com/article/10.1007/s00502-018-0635-2)

The speed at which images can be developed is affected by the speed at which it can be scanned into the system. A variety of scanning methods are available for different purposes such as azimuth and elevation, dual oscillating plane mirrors, dual axis scanner and polygonal mirrors. They type of optic determines the resolution and range that can be detected by a system.

Photodetector and receiver electronics

Sensors which convert optical power to electrical power, using the photoelectric effect, are called photodetectors.

Leave a Reply

Your email address will not be published. Required fields are marked *