Using dToF in LiDAR applications
LiDAR stands for ‘light detection and ranging’ and is a technique for measuring the distance of objects from a sensing device. The principles used are very similar to RADAR but, with LiDAR, the radio waves are replaced by light – usually laser light. The LiDAR system emits a beam of light that hits a target before being reflected back to a sensor that is located close to the light source. By measuring the time taken for the light to travel, and knowing the constant speed of light, the distance of the target can be calculated with a high degree of accuracy. By repeating this measurement at regular intervals across a scene or target area, a 3D map of the environment can be built up.
LiDAR has applications in the automotive world, particularly in the advanced driver assistance system (ADAS) where it can be used for obstacle detection and collision avoidance as well as adaptive cruise control (ACC) and navigation. However, while the automotive uses are frequently talked about, they represent a very small proportion of the total applications for LiDAR.
In space exploration, LiDAR can be used to create 3D topographical maps of the surface of planets, or to calculate the distance between a celestial body and Earth. Agriculture uses LiDAR to survey fields and crop conditions, allowing farmers to model and predict crop yields as well as monitoring crop growth.
In forestry, applications include measuring forest canopies and monitoring deforestation as well as being used in the proactive management of forest fires. Environmental conservation can also benefit as LiDAR is used to manage coastal erosion and monitor dunes as well as collecting data on glacier erosion. During natural disasters, LiDAR can be used to predict Tsunamis or to rapidly evaluate damage from earthquakes.
There are a multitude of industrial applications, including inspection on the production line in factories or planning large construction projects such as buildings or roads. Another use is protecting danger zones, such as those around railroad crossings. LiDAR can guide robotic vehicles safely around facilities which has applications in warehousing, docks and airports, to name just a few.
In fact, there are literally hundreds of uses for LiDAR in areas as diverse as transport, architecture, surveying, gesture recognition, mining, law enforcement, renewable energy and geology.
The use of dToF in LiDAR applications
The key principle behind LiDAR is direct time-of-flight (dToF). In a typical LiDAR system, a laser is used to produce pulses of light. When this light hits an object in its path, light is reflected and, while much of the light is scattered, some will be reflected back to the sensor in the LiDAR system.
An accurate clock within the system is able to determine the time taken for the light to reach the object and return. As the speed of light is a constant (c), the distance of the object can easily be calculated. With a very accurate clock, very high levels of precision are obtainable with LiDAR systems.
While knowing the distance to a point is useful, by moving the light source in a scanning pattern and recording each distance value, along with its position in the scan, a 3D map can quickly be built. This is the primary use of LiDAR and facilitates many of the emerging applications mentioned earlier in this article.
Silicon PhotoMultiplier (SiPM) – the essential sensing element
The ability to accurately capture and quantify the reflected laser light is critical to the performance of any LiDAR system. One of the best performing sensors for LiDAR systems is the Silicon PhotoMultiplier (SiPM) which integrates a dense array of small, independent Single Photon Avalanche Diodes (SPAD) sensors.
Each of these minute sensing sites includes a quench resistor and is known as a ‘microcell’ which is just a few microns square. When a microcell absorbs a photon, a Geiger avalanche may be initiated causing a large photocurrent to flow through the microcell. This, in turn causes a voltage drop across the quench resistor which reduces the diode bias and quenches the current, thereby preventing further Geiger-mode avalanches. The microcell then resets, ready for the next measurement.
Typically, a SiPM will have between 100 and several thousand microcells per square millimetre, each of which detects photons identically and independently. The current from each microcell is summed to form a quasi-analog output which gives information on the magnitude of an instantaneous photon flux.
Alternate techniques for detecting and measuring photons include avalanche photodiodes (APD) and PIN diodes. However, SiPM sensors are an improvement over both of these due to their ability to detect single photons and their high gain. This enhanced performance allows the detection of low reflectivity targets at a very long range, as is required in many LiDAR applications.
SiPM sensors are available as fully packaged solutions, such as the RB-Series from ON Semiconductor that is sensitive to the red and NIR region of the electromagnetic spectrum. All of the sensors in the series feature high responsivity, fast signal response, low operating voltage as well as a low temperature coefficient of operating voltage.. They are packaged in a small (1.5 mm x 1.8 mm) and robust MLP (molded lead frame package) suitable for solder reflow processes.
There are three types available, distinguished by their microcell size (10 ìm, 20 ìm or 35 ìm) although each device has an active sensing area of 1 mm x 1 mm. The larger microcell varients offer higher detection efficiency, , whereas a smaller microcell gives a higher dynamic range . All devices have a fast signal response of around 1.0 ns for the standard output and around 500 ps for the fast output.
SiPM dToF LiDAR Platform
While the principles of dToF are relatively simple, developing a fully-functioning solution can be challenging due to the accuracy with which the SiPM detects the returned light. In order to support engineers wanting to develop high-performance LiDAR industrial range finding applications, ON Semiconductor offers a development platform with a complete LiDAR solution based on the dToF principle.
Intended to reduce development efforts, the platform is practically a turnkey solution with software adjustable settings for multiple applications. The overall system cost is optimized and all hardware (BoM, schematic, PCB Gerber), software and source files are provided.
The system offers dToF capability for a single point within the range 1 mm to 23 m and includes plano-convex lenses to maximize the distance measurement. Only a single 3.3 V or 5 V power supply is required for operation. A dedicated GUI is included to configure frequency and pulse width as well as to set the buck and boost voltages.
While the principles of dToF LiDAR are simple, implementation can be challenging, especially for first-time designers. ON Semiconductor® provides a reference platform to demonstrate how effective LiDAR can be when using the best technologies. The comprehensive platform provides all of the information, hardware and software that designers will need to get to proof of concept quickly and reliably.
About the author:
Edel Cashman is Senior Applications Engineer, ON Semiconductor. Having trained in microelectronics, Edel previously worked in analog IC design and biophotonics design where she first came across SiPMs. She joined the team at SensL in 2015, and SensL was acquired by ON Semiconductor in 2018. Edel is expert in SiPM technology and detector design, working daily on a wide range of applications, with leading LiDAR developers worldwide.