mmWave micro-radar for situational awareness & drone navigation
Author : Aled Catherall | Chief Technology Officer | Plextek
01 October 2022
Unmanned aerial vehicles (UAVs – also known as drones) have taken off in a big way, from low-cost hobby use through package-delivery trials & filming entertainment or sporting events to front-line military use in Ukraine.
This article was originally featured as the cover story in the October 2022 issue of EPDT magazine [read the digital issue]. And sign up to receive your own copy each month.
By 2022, the global UAV market is forecast to reach more than $21 billion and drive a market economy for business services valued at over $127 billion. Industries that will be rapidly disrupted by drones include infrastructure, agriculture, transportation, defence and security, entertainment and media, insurance, telecommunications and mining. Here, Aled Catherall, Chief Technology Officer at RF-focused design, engineering & technology consultancy, Plextek assesses the potential of millimetre-wave radar in a form compact enough to deploy on a small drone to provide sense-to-avoid data…
While drone technology has advanced rapidly, how to navigate and avoid collisions remains a key challenge. Currently, in the absence of an exemption licence, drones must be piloted, or at least have human pilot back-up, and flown within the line of sight of that pilot. Regulation also limits the use of drones in built up and uncontrolled areas – and such regulation is necessary to avoid collisions with obstacles such as power lines, trees, lampposts, other drones and, of course, people.
Finding the right sensor technology to detect obstacles is key if the full potential for unmanned drones is to be realised. Camera systems are in widespread use, as sensors that provide information about the surrounding environment. Modern deep learning algorithms, such as semantic segmentation, can also interpret images of complex and cluttered scenes, providing means to accurately identify and locate potential hazards. The topic of vision-based navigation is currently subject to significant academic and industrial research, and capability here is accelerating rapidly.
Plextek_mmWave micro-radar for drone navigation
However, while recent advances in vision-based navigation are encouraging, there remains the problem of lighting and resolution. Performance using optical cameras will tend to degrade in low light conditions, and whereas short-wave infrared can alleviate some of the issues, enabling better operation in the dark, the resolution is likely to be too poor to spot fine obstacles such as power lines or wire fences at sufficient range to plan an avoiding route.
Lidar (from light detection and ranging), a detection technology system which uses light from lasers, has its attractions – but can struggle to detect very dark objects or smooth ones at specular angles, including puddles of water or large glass panes. It is also usually adversely affected by rain or snow precipitation, as well as fog or smoke. In addition, its performance depends on the amount of ambient light, usually performing better at night than in the day.
The mmWave radar approach
There is a good alternative or complimentary technology to camera or Lidar. Radar (from radio detection and ranging) operating in the 60 GHz band offers detection ranges of many tens of metres, and is able to detect very small targets, regardless of the time of day or adverse weather, including fine targets such as power lines.
At Plextek, we have engineered a millimetre-wave (mmWave) radar that incorporates a fast electronic scan and the supporting electronics to achieve real-time data capture and processing. The 60 GHz band takes advantage of technology advances in consumer telecoms, such as reduced antenna size, as well as LPI/LPD (Low Probability of Intercept/Low Probability of Detection) analyses. This also means that scanning radar systems can be based on low-cost COTS (commercial off-the-shelf) SiGe (silicon-germanium) chipsets, with full radar functionality on a single PCB (printed circuit board).
A high-resolution mmWave radar may lack the fine detail that a camera creates, but it much more readily reflects the geometry of the world it views, since it directly provides range estimates to each object in the scene. Our radar also measures Doppler – a shift in frequency caused by motion – and this can provide a rapid estimate of how quickly something is moving toward the drone to support avoidance measures. A radar that scans in both azimuth and elevation so provides a direct, 3-dimensional estimate of its location.
Results from tests we have conducted using our radar demonstrate successful detection of small, static obstacles above ground clutter, such as power lines, metal poles, people and cars, at sufficient ranges to enable evasive action and route planning. In addition, Doppler processing enables discrimination of moving objects, such as other small drones or birds.
Radar is widely used across many industries; for instance, in automotive, to determine the relative position and motion of vehicles on the road, to operate adaptive cruise control or trigger the deployment of airbags before an impact. But with a few exceptions, these radars do not scan – they stare with fixed beams. More advanced mmWave scanning radars have the potential to elevate applications such as driverless cars and UAVs to new levels.
More information...
Contact Details and Archive...
