As the market for drones continues to grow there is an increasing need for accurate sensing and obstacle detection technologies. LiDAR is a strong candidate as it allows an accurate 3D map of the drone's surroundings to be sensed, shows excellent performance in poor weather conditions and can be easily integrated onto drones.
We spoke to Pier-Olivier Hamel from LeddarTech, a LiDar sensor company, on how LiDar sensing is going to become a key technology for autonomous drones and the benefits of their industry leading Leddar sensors.
Why do you think that LiDAR sensing is a good fit for drones and other UAVs?
LiDAR is a very versatile technology, providing benefits that either complement other detection and ranging methods, or just plainly outperforms them.
Recently, LiDAR has gained a great deal of traction among drone and unmanned vehicle manufacturers thanks to its many advantages over traditional sensing technologies, such as radar and sonar. Indeed, LiDAR has much more flexibility when it comes to its specifications, integration and form factor.
Other technologies in use today for drone applications are ultrasound and stereo cameras, both of which have a very limited range This is not the case for LiDARs. Although stereo cameras can create very high resolution images, they are more sensitive to extreme or changing lighting conditions.
That being said, it is important to mark the distinction between scanning and solid-state LiDARs. Scanning LiDAR is better known for being the technology that enabled revolutionary applications, such as Google self-driving cars or high-resolution terrain mapping. Scanning LiDAR systems are capable of generating very precise renders of their surroundings, albeit at a great cost. Indeed, scanning LiDARs are typically very expensive, bulky, heavy, and require fragile mechanical mobility: this exact combination of factors makes them hardly viable for integration into commercial UAVs.
Solid-state or fixed-beam LiDARs differ from scanning LiDARs: the light source that serves as the detection medium is emitted in a fixed, diffused beam rather than being projected in all directions as single-point rays. Traditionally limited in range and resolution compared to their scanning counterparts, solid-state LiDARs have the advantage of requiring much less processing, in addition to being a lot smaller and lighter.
Solid-state LiDARs, as the name implies, have no moving parts, which provides great robustness and durability. For all of these reasons, solid-state LiDARs tend to be vastly more affordable than scanning LiDARs, albeit lacking the range, resolution and sensitivity to make them a top candidate for drone applications.
LiDar sensors are one of the key technologies used by drones to perceive their environment and avoid collisions. Shutterstock | Dmitry Kalinovsky
What is the difference between LeddarTech’s sensors and the LiDAR sensors produced by your competitors?
In its core, Leddar technology has a higher sensitivity and better immunity to undesirable signals as well as other benefits from its powerful digital signal processing algorithms.
This means that Leddar devices, in addition to being fully solid-state compared to mechanically scanning LiDARs, are more reliable, better equipped to deal with harsh conditions and direct sunlight, and are more cost effective.
Combining the form factor, richness of data and price range suitable for the development of commercial UAV autonomous systems, LeddarTech has developed a direct time-of-flight approach that brings together the best of both scanning and fixed-beam LiDAR sensing, and improves it by a few notches.
Rather than working directly on the analog signal, as traditional LiDAR does, the patented Leddar light processing technology iteratively expands the sampling rate and resolution of a signal using multiple successive light pulses. It then analyses the resulting discrete-time signal to recover the distance for every object present in its field of view.
The complete digitized waveform produced by this method holds lots of information that can be processed. This rich data input results in more reliable measurements in a wider range of conditions than traditional time-of-flight methods, resulting in a much higher effective dynamic range.
As Leddar sensors are fully solid-state LiDAR sensors they are more reliable than competing technologies, allowing them to have a more accurate perception of their environment.
How easy is it to integrate Leddar sensors into a drone?
Integrating Leddar sensors into a drone is fairly simple.
For example, any embedded platform will be able to “speak” with a Leddar module using a simple serial port. LeddarTech also provides software development kits to facilitate their devices’ integration by drone manufacturers and integrators.
The flexible and modular form factor of the new generation of Leddar sensors such as those developed from the LeddarVu platform also greatly facilitates physical integration into all kinds of drones and UAV bodies.
What applications can LiDAR sensing enable for drones and UAVs?
Since LiDARs can be used for navigation, the applications are extremely varied, from pilot assistance and collision warning to higher levels of autonomy in any type of drone.
LiDAR sensors are also capable of altimetry for smooth takeoff or landing. The LeddarOne rangefinder module is very popular for this kind of application, already being used in commercial drone models. It has a fairly long range for its power consumption, and its 2-inch diameter and low price makes it a very well rounded solution for drone manufacturers.
One of the first UAV manufacturers to use Leddar technology is Draganfly Innovations Inc., which integrated the LeddarOne rangefinder as a standard feature on their Draganflyer Commander model. The sensor is fully integrated into the autopilot system which works in conjunction with a precision barometric sensor to provide the best in altitude performance.
Another example is Robota, who chose the LeddarOne for its low pricing, compact size, superior power and range, and for the ease of doing business with LeddarTech as a supplier.
LiDar sensors can be used to detect the altitude at which a drone is flying.
How far away do you think we are from drones that have a full, 360° perception of their environment? What technology needs to be developed to realize this?
As of today, there are currently no fully autonomous systems that can provide reliable multi-directional sense-and-avoid capabilities for commercial UAVs.
Such systems should be capable of detecting potential obstacles, assessing the risk of collision and taking evasive actions all by themselves, overriding the pilot’s input if needed, in a 360° sphere around the device.
That being said, I think it is fair to assume that 360° perception on commercial and consumer drones is probably still a few years away. Beyond the sensors, a lot of core functionalities need to be perfected to achieve that, like perception algorithms and artificial intelligence, all at a price and weight suitable for these drones.
Can you see LiDAR sensors being used to develop autonomous drones?
Yes, LiDAR will likely be a key component of autonomous drones, along with multiple other sensors and components.
Very much like for autonomous cars and road vehicles, fully autonomous drones – including reliable security and collision avoidance systems – will use a sensor fusion approach that will include LiDAR sensing.
LiDAR technology will allow drones to have a complete 360° perception of their environment.
Does LeddarTech offer custom solutions to drone manufacturers?
Yes, LeddarTech can offer custom solutions to drone manufacturers who need a solution tailored to their own applications and needs.
We can support any project to some degree, including software, mechanical and optics design, integration, and more. It all depends on the application.
Where can our readers find out more about LeddarTech’s solutions for drones and UAVs?
More information on LeddarTech’s drone and UAV solutions can be found on our website.
Download LeddarTech's Unmanned Vehicle Magazine
About Pier-Olivier Hamel
As Product Leader at LeddarTech, Pier-Olivier Hamel manages all custom projects developed by the company, thanks to Leddar technology’s open and flexible approach to system integration.
He has been with the company since 2013, notably as Applications Engineer. Specializing in embedded systems design and telecommunications, Pier-Olivier formerly worked as product specialist, software developer and product designer for other companies.
Pier-Olivier holds a bachelor’s degree in Electrical Engineering from Laval University in Quebec City.
Disclaimer: The views expressed here are those of the interviewee and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.