Audi 's the self-propelled unit approached the launch with a single access to Lidar while testing in Munich using a fleet of autonomous electrical electronic crossovers.
The Audit-Intelligent Ride, or AID subsidiary, said on Wednesday that it uses leverage sensors developed by Aeva, founded two years ago by Apple and Nikon veterans.
Aeva, The California-based Mountain View, launched by Soroush Salekhan and Mina Ressk, has developed what it describes as a "4D lidar", which can measure distance as well as current speed without loss of bandwidth, while preventing mixing from the sun or other sensors. Moving past 4D branding-talking, and technology is huge.
Lidar, or radar for detecting and measuring light, measures the distance. Many (with Tesla as one exception) consider it to be critical and indispensable sensors in the new automated driving. For years, Velodin dominated the industry.
Today there are dozens of leading startups that have emerged with promises of technological breakthroughs that will offer sensors at a lower price with better resolution and accuracy than Velodine. It is a promise that is filled with challenges, especially the ability to expand production.
Traditional lidar sensors can determine the distance by sending high-power impulses to the light outside the visible spectrum, and then tracking how much time it takes for each of these impulses to return. As they return, the direction and the distance to, regardless of the fact that they have affected these impulses, are recorded as a point and eventually forms a 3D map.
Aeva sensors emit a continuous low power laser, which allows them to experience fast speeds at each point in the frame of up to 300 meters, the company said. In other words, the Aeva sensors can determine the distance and direction, as well as the speed of objects that come or move away from them.
This is a practical opportunity for perceptions of autonomous vehicles operating in an environment of objects traveling at different speeds, such as pedestrians, bicycles and vehicles.
Aeva, supported by investors, including Lux Capital and the Canaan Partners, says its sensors are unique because they are "free" from interference from other sensors or sunlight.
It was this combination of long-range perspective, instantaneous measurements of the speed at precision cm / s and robustness of obstacles that sold AID CTO Alexandre Haag to Aeva sensors.
Aeva lasted 18 months through the validation process with Audi and the parent company Volkswagen. This announcement confirms that Aeva made the past a critical hurdle in Audi's AV-plans. Aeva sensors are already on the Audi e-tron development vehicles in Munich. The automaker plans to bring autonomous driving to urban mobility services over the next few years.
Interference is possible and can cause a stream of random spots on a 3D map if the lid is pointed directly into the sun or if there are more sensors on the same vehicle. Lidar companies have implemented various techniques to prevent interference patterns; autonomous vehicle developers also anticipate potential interference with sun and snow problems by creating algorithms for rejecting these types of outliers.
However, Salekhan argues that interference is a significant challenge.
When talking about the challenge of building a scale and designing at a mass scale, it's not just about how easy it can be produced, says Salekhan. "It's also about the fact that these things work together in a uniform order. So when you talk about hundreds of thousands of cars, that's a big deal."