All weather perception for self-driving cars
RADAREYE is a startup focused on commercialising recent breakthroughs in cross-modal radar-camera self-supervised learning.
The company is targeting the automotive market as an immediate route to the commercialisation of its technology, with medium-term plans for home automation, healthcare, smart cities, and industrial surveillance applications.
Radium I is our first version of all weather perception stack for self-driving.
Our competitive analysis shows how Radium offers an improvement with respect to existing approaches in terms of reliability (works in bad weather), form factor, cost, and scalability. This differentiation is summarised as follows.
| Improvement | Description |
|---|---|
 All weather |
Works in all weather—fog, rain, snow, etc.—whereas other solutions will not. |
 Light & compact |
Hardware needed for Radium is much smaller than alternative lidar hardware. |
 Cheaper hardware |
Hardware needed for Radium is much cheaper than alternative lidar hardware. |
 Data scalability |
Our deep learning model does not need to train on human labelled data. Instead, it automatically trains using pairs of radar and image data. Therefore, Radium is vastly more scalable. |
Radar is not susceptible to the same limitations as cameras — it can see through bad weather like fog and detect cars.
We have built an ML model that is able take the radar data and recognise the patterns that represent cars.
The output of our model is the position of cars. This inference is used within car navigation system.
Figures show how we train Radium (on the left), and how we deploy Radium (on the right). Foggy images are the reference visual scenes of the radar heatmaps on the right. Radar allows us to accurately detect and localise cars under bad visibility, whereas detection is not viable from the corresponding foggy images.
RADAREYE’s co-founders have 27+ years combined experience in silicon, radar, perception, and automotive, most recently at Nokia Bell Labs and General Motors. The company is based in London.
This is bold and this is strong. This is italic and this is emphasized. This is superscript text and this is subscript text. This is underlined and this is code: for (;;) { ... }. Finally, this is a link.
Fringilla nisl. Donec accumsan interdum nisi, quis tincidunt felis sagittis eget tempus euismod. Vestibulum ante ipsum primis in faucibus vestibulum. Blandit adipiscing eu felis iaculis volutpat ac adipiscing accumsan faucibus. Vestibulum ante ipsum primis in faucibus lorem ipsum dolor sit amet nullam adipiscing eu felis.
i = 0;
while (!deck.isInOrder()) {
print 'Iteration ' + i;
deck.shuffle();
i++;
}
print 'It took ' + i + ' iterations to sort the deck.';
| Name | Description | Price |
|---|---|---|
| Item One | Ante turpis integer aliquet porttitor. | 29.99 |
| Item Two | Vis ac commodo adipiscing arcu aliquet. | 19.99 |
| Item Three | Morbi faucibus arcu accumsan lorem. | 29.99 |
| Item Four | Vitae integer tempus condimentum. | 19.99 |
| Item Five | Ante turpis integer aliquet porttitor. | 29.99 |
| 100.00 | ||
| Name | Description | Price |
|---|---|---|
| Item One | Ante turpis integer aliquet porttitor. | 29.99 |
| Item Two | Vis ac commodo adipiscing arcu aliquet. | 19.99 |
| Item Three | Morbi faucibus arcu accumsan lorem. | 29.99 |
| Item Four | Vitae integer tempus condimentum. | 19.99 |
| Item Five | Ante turpis integer aliquet porttitor. | 29.99 |
| 100.00 | ||