The origins of LiDAR (“lidar”)
The genealogy of lidar (light detection and ranging) stems from sonar (sonic navigation and ranging) and radar (radio detection and ranging) technologies. Sonar and radar use acoustic (sound) and electromagnetic waves in the radio or microwave frequency spectrum, respectively. Lidar technology uses electromagnetic radiation in the infra-red and visible light spectrum, which are much higher frequencies than that of radio or microwaves.
The primary purpose of the three technologies is to detect and measure the distance to objects or surfaces through wave reflection and detection. Each system uses a high-accuracy timing device to measure how long it takes from the time that the emitted energy, be it sound or electromagnetic radiation, to travel both to and from the object or surface. Using this duration and the speed of the sound/radiation, the range can be accurately calculated.
- Sonar is typically used for subsea applications due to acoustic waves being able to travel long distances through water. Note that radar cannot penetrate any useful distance through water.
- Radar is typically used for airborne applications, such as air traffic control, defence, weather forecasting, and even detecting objects underground. Radar systems are typically large, expensive and specialised, which generally removes them from the realm of common commercial use.
- Typical lidar applications include topological and vegetation survey and mapping. It is also very widely used for inspecting utilities such as high-voltage power lines, dams, water supplies, buildings, and road infrastructure. As with many technologies of the current era, the speed at which an idea transforms into smaller, faster and cheaper devices is continuing to accelerate. This is a fundamental reason behind why lidar is so popular – to the point of it becoming small enough and cheap enough to be a standard gadget on some of the latest smartphones.
Laser is the key to lidar technology
Lidar uses laser technology to emit light (photons) that has the necessary wavelength to suit the desired purpose. The beauty of laser technology is that it is capable of emitting light with extreme consistency; that is, in terms of its wavelength, polarisation and direction, over time. These properties enable the laser beam to have a very narrow range of wavelengths, giving it a distinct, singular (monochromatic) colour (if visible light) and for it to be very tightly focussed, which creates the classic long and narrow beam of light, rather than one that spreads out in all directions.
Laser wavelength is inversely proportional to the frequency and, because the speed of light in any given medium is constant, knowing one means you can determine the other. In a lidar application, the desired propagation and reflective properties of the laser will determine the wavelength / frequency to use. For example, if the lidar is being used to detect the particle concentration of a specific gas in the atmosphere, the laser wavelength must be reflective of the target particles.
The power of the laser is also very important in lidar surveying applications and can be correlated to the maximum distance away from the lidar that objects can be detected. The higher the output power of the laser, the farther away the lidar can be from detectable objects. In UAV-lidar surveying, a higher-powered lidar scanner can be used at greater distances AGL (above ground level) / altitude. Increases in AGL increase the swath width of the scanned area for each pass, making for faster scanning of the survey area. Note that laser power in typical lidar surveying applications is limited by eye safety regulations.
Image depicting the increase in swath width when UAV-lidar surveying at higher altitudes, enabled through using higher power lidar scanning lasers. An additional benefit is that the UAV can be used at altitudes that make them less prone to collisions with trees and power lines, for example
The extremely short wavelength of a typical lidar laser makes the actual beam itself a very small diameter. This small size enables a lidar scanner to produce very high-resolution / high density data due to the sheer number of laser emissions and reflections that it can produce. The generated data, which is essentially the range information for each emission/reflection, is known as a point cloud.
What is driving the lidar surveying industry?
Our hunger for data has never been stronger than it is now. The quest to acquire data for research, knowledge, and understanding is spurring many great advances in sensors and other technologies that allow us to detect and measure things remotely and in automated ways that not so long ago were limited purely to the imagination.
Lidar is such a technology. Lidar allows us to capture information about a site, object or landscape that can be represented in a highly visual way – a point cloud map, 3D mesh or model that we can analyse to understand and estimate various conditions and characteristics of the survey site. A point cloud may contain many, many millions of reflected laser points, with each point accurately referenced to absolute position (longitude and latitude coordinates) on the Earth’s surface. This mosaic of points can produce a very detailed and accurate digital twin of the survey site. Point cloud maps and models will typically apply colour gradients to represent variations in range of the reflected light to the scanner, making human interpretation very intuitive. This makes the output of lidar surveying invaluable to clients that, without a readily understandable graphical model, may be otherwise unable to interpret the data.
Example lidar survey point cloud. Image courtesy EyeVi Technologies
Something quite peculiar to lidar is that a laser emission may be able to have multiple reflections, depending on survey site conditions and angle of laser penetration. A good example of the benefits of this over, say, photogrammetric (photographic image based) surveys, is that lidar is able to filter through vegetation to an extent, in much the same way that light filters down through a forest canopy to the forest floor, and is capable of receiving multiple reflections from the same emission. That is, some photons will reflect off part of a leaf, for instance, others a branch, and some photons bouncing off the ground surface. The amount of energy in the reflected light (the number of photons or the intensity of the reflected light) usually determines some property of the object or surface; for example, its opacity, shape, material and absorption of light. With our forest example, this means that distance to the ground surface can be measured as well as the distance to treetops and objects in-between, making lidar an exceptional tool for topographical survey of densely vegetated areas.
Example image shows how a laser emission passing through a tree may provide multiple reflections as parts of the laser beam encounter various objects, including the ground surface
Why is UAV-lidar popular for surveying?
In the interests of keeping in theme with small-scale and more affordable commercial applications of lidar surveying, we will ignore lidar surveying from piloted aircraft and concentrate on UAVs (uncrewed / unmanned aerial vehicles) or “drones”.
The UAV has been around a lot longer than might be imagined, however, since 2006, UAVs entered the commercial space and a few years later were introduced into the personal market. The explosion of UAV drone use since, which some describe as a “drone revolution”, has been instrumental in making lidar surveying what it is today. The fact that a drone can typically travel in any direction as well as hover with high levels of control and stability makes it extremely well suited to surveying and inspection both in, around, and under structures as well as from above.
It is fairly safe to say that the evolution of lidar, since it first appeared in the 1960s, has accelerated since GNSS (global navigation satellite system) receivers and inertial measurement units (IMUs) became commercially available in the 1980s. GNSS and IMU technology enables high accuracy satellite-based navigation and a detailed understanding of vehicle motion that, together, opens up a new world of possibilities for the UAV. If you couple UAV drones and inertial navigation systems and lidar, the result is a vehicle that is easily and relatively cheaply deployed that can provide outstanding results for mapping the surroundings, georeferenced to absolute position on the Earth’s surface.
The cost of lidar scanners is reducing, as is the physical size and weight as the technology further advances. In conjunction with this is increasing scanning speed and the resulting survey resolution. Currently, the fastest scanners can operate at a staggering two million points per second and even more, providing exceptionally detailed point clouds. Similarly, the ranging distance of lidar scanners is increasing, allowing the scanner to be used at higher altitudes or further from the survey site whilst still producing acceptable results.
The speed at which high-resolution UAV-lidar surveying can be carried out and its versatility has many benefits over traditional surveying techniques, such as:
- Cost – reductions in cost over using human surveyors and traditional techniques cannot be overstated. Being able to survey remotely or even autonomously without having to put surveyors’ boots and equipment on the ground or charter aircraft, for instance, presents massive reductions in costs and logistics.
- Speed – a combination of equipment that can be easily handled and deployed that is capable of quickly and accurately covering large areas greatly speeds up survey time. Post-processing of acquired data makes it possible to deliver survey results to clients within hours.
- Safety – removing the need for personnel to perform surveys on-site keeps them out of harm’s way, especially in areas with poor access or hazardous terrain. This extends to other high-risk survey sites; for example, inspection at altitude and high-voltage installations.
- Survey frequency – lower cost and simple deployment brings opportunity for more frequent surveying. This can enable faster detection of environmental changes, damage or deterioration and provides opportunity to respond in a more timely manner.
- Environment – UAV-lidar presents a more sustainable and environmentally friendly option over previous techniques that typically involve multiple vehicles, more personnel, higher carbon emissions, and greater potential for environmental damage.
- Commercial opportunity – UAV-lidar continues to be an excellent business and commercial platform due to its relatively low cost and its increasing importance in the burgeoning survey and inspection markets.
Other practical lidar surveying applications
Applications for lidar continue extending far beyond aerial topographical survey. In fact, if there is anything that requires visual inspection, it is quite possible that a lidar based solution is available or likely will become available.
The beauty of lidar technology is that it lends itself to a tremendous array of applications that have traditionally required very large amounts of labour and time and associated costs. Below are some examples:
- Rail corridor surveying – train-mounted lidar that can measure track gauge and alignment, ballast distribution and coverage, and clearance between potential hazards to the train, for example.
- Road surveying – vehicle-mounted lidar that can be used as part of an automated system that assesses road conditions and identifies and georeferences areas that require maintenance and the type and extent of action required.
- Town and infrastructure planning – simplified, fast and cost-effective UAV-lidar surveying improves the ability to make informed town planning decisions and assists in other municipal tasks that require an understanding of the environment at particular locations, such as parks and recreation sites and when zoning land for development.
Utilities inspection – difficult and hazardous inspection tasks, such as dam walls, high-rise building construction sites, bridges, and high-voltage power lines, insulators and pylons can be rapidly completed from the safety of the ground. UAV-lidar can provide detailed imagery that can be quickly analysed and processed. The savings made by using UAV-lidar in terms of reducing crew numbers and the use of specialised vehicles is vital to sustainable and repeatable inspections.
- Precision agriculture – precision agriculture practices rely heavily on multiple sensor technologies to provide farmers with deep insights into the state of the farm and land in general. Lidar attached to UAVs can provide high accuracy location data for identifying pests, disease, weeds, and fence damage, for example. Lidar mounted to ground vehicles or robots can be used to count fruit on trees, and assess crop condition amongst a growing array of tasks that can be automated or simplified, leaving the farmer to concentrate on running the farm and no longer needing to perform some of the more menial activities.
- Coastal bathymetry – advances in lidar are making it possible as a surveying technology for shallow water aerial bathymetry lidar (ABL). Instead of waterborne vessels and traditional sonar-based bathymetric mapping, in the right conditions, aircraft or UAV mounted lidar can be used that maps not only the surface of the seabed, but the coastal fringe simultaneously. The speed at which ABL surveying can be conducted is many times faster than traditional sonar bathymetry, making the benefits obvious.
Artificial intelligence and machine learning augment lidar technology
The emerging autonomy revolution and the idea of automating tasks that are labour-intensive, hazardous or otherwise inefficient for humans to perform, is driving deep innovation in many diverse fields and applications. The notion of combining lightweight and uncrewed aerial vehicles with satellite navigation, lidar and inertial navigation systems continues to provide excellent tools for survey and inspection that is hard to beat, so what improvements might we see?
Artificial intelligence (AI) and machine learning (ML) software are technologies that can be used with lidar to great success. As lidar resolutions increase, the ability for object identification and classification also goes up. AI/ML software is typically “trained” in the lab to recognise objects, creatures or phenomenon by characteristics of shape and size. Moreover, complex automated analysis is becoming possible with increased lidar point cloud resolution that enables more complex shapes and smaller sized objects to be recognisable.
New and improved remote sensing technologies are rapidly increasing the spectrum of data acquisition. For example, hyperspectral imaging of atmospheric and ground-based chemical composition for agricultural use. By fusing information from third-party sensors with lidar and navigation data, the resulting survey point clouds can include a gamut of sensor data that makes analysis of multiple phenomena possible. It is also becoming more common to gather multiple data types in a single survey rather than multiple surveys using different sensing equipment, which saves time, reduces cost to clients and minimises emissions.