General Motors has invested millions of dollars in Oculii, a US-based startup that tries to improve the performance of radar sensors for use in automated driver assistance systems (ADAS) and autonomous driving. The news comes shortly after Oculii’s funding round in May 2021, where it raised $ 55 million from a range of OEM and Tier 1 vendors, including Hella. The startup already holds around ten patents in the field of radar detection.
Radar sensors emit a focused beam of radio waves in front of the vehicle to detect obstacles by reflecting that beam back to a receiver. This configuration provides exceptional accuracy for distance detection and works in almost all weather conditions. Additionally, radar sensors are more affordable than competing LiDARs (light-based sensors) and alternative approaches. Radar is an ideal sensor for autonomous Level 1 systems such as adaptive cruise control which can match the speed of the car in front.
However, the “resolution” of the image provided by the radar is poor. While the radar can indicate very precisely how far an obstacle is from the vehicle, it cannot tell what type of obstacle it is or in what direction it is moving, other than its proximity to the radar sensor. This means that the application of radar sensors in Autonomous Systems Level 2 and above is limited as they have to create a more detailed picture of the world around them – including the ability to differentiate between static obstacles, other vehicles. and pedestrians, and their relative trajectories.
Most Level 1 and Level 2 stand-alone systems sold today include radar as part of the overall sensor suite, alongside existing vision cameras, ultrasonic and LiDAR sensors. Tesla, however, stands out for its decision to omit the radar from Model 3 and Model Y vehicles built from May 2021, relying instead on its so-called Tesla Vision system based entirely on vision cameras and processing. ‘images by machine learning. The company insists the system can collect enough detail from its vision cameras to safely categorize and determine the location of obstacles around it, and chart a safe course forward. However, critics have pointed out that vision cameras can be confused by bad weather or unexpected scenarios.
Oculii believes it can make radar sensors a more attractive option for Tier 1 manufacturers and OEMs by using software to improve the resolution of existing commercially available radar sensors. His system uses artificial intelligence to dynamically adjust radar waves in response to the environment. While Oculii is understandably secretive about exactly how this is done, he claims the software can improve the detection accuracy of existing radar systems up to 100 times.
Oculii’s demo videos show a vehicle equipped with at least four of its radar sensors moving through a busy road environment, with a detailed heat map of the location of obstacles using its radar sensors. By combining four sensors at the front, rear and sides of the vehicle, it is possible to achieve 360-degree radar vision up to 350 meters, regardless of the weather.
The company does not plan to manufacture radar equipment, although it does offer a small range of radar sensors to complement its products. Its business strategy going forward will be to focus on developing its software solution which, crucially, can be applied to existing radar sensors already available on the market. This will reduce the time and R&D effort required to integrate the Oculii system into an existing ADAS or standalone software package.
General Motors is already rolling out its Level 2 autonomous system dubbed Super Cruise on models such as the Cadillac Escalade and Chevrolet Bolt EUV, and plans to offer it on 22 different models by 2023. It is likely that a future iteration of Super Cruise will incorporate Oculii radar technology to improve the accuracy of its sensors and overall safety.