“Lidar is a fool’s errand,” Elon Musk said. “Anyone relying on lidar is doomed. Doomed! [They are] expensive sensors that are unnecessary. It’s like having a whole bunch of expensive appendices. Like, one appendix is bad, well now you have a whole bunch of them, it’s ridiculous, you’ll see.”
However, not all LiDARs have equal opportunities for development. In fact, last week it was reported that Audi chose the Lidar sensor from Silicon Valley startup Aeva for its autonomous driving program. At present, there are two main types of LiDAR technology on the market: one is the technology using amplitude modulation (AM) method, and the other is the new LiDAR technology using frequency modulation (FM) method.
Companies using AM LiDAR modulate the amplitude of the pulses in a rotating laser array and then calculate the time it takes for the light to hit the object and bounce, the time-of-flight (ToF) technique. It uses this information to achieve target recognition within the perceived area, such as other cars or pedestrians.
By far, more than 95% of the $1.1 billion "Lidar Bubble" has been invested in companies that are pursuing AM sensing.
According to Aeva in Silicon Valley and Blackmore in Montana, they developed another sensing method for lidar. Blackmore spent more than 10 years developing FM LiDAR for the Department of Defense, whose LiDAR sensor was used for border scanning and intrusion detection. It began selling products to original equipment manufacturers (OEMs), Tier 1 suppliers and autopilot (AV) startups around the world in early 2018. At the same time, it is favored by some powerful companies and supported by two major industry giants: AI Ventures and BMW i Ventures.
Aeva and Blackmore modulate the frequency of the laser beam rather than the amplitude. The FM LiDAR does not emit a pulse wave like the AM LiDAR. Instead, it produces a continuous wave with a small frequency change. LiDAR measurements utilize the Doppler effect, that is, when the source and observer are facing or apart from each other, the frequency of the wave increases or decreases accordingly.
According to Doppler lidar proponents, traditional AM lasers are very susceptible to interference from sunlight and other sensors. It is computationally intensive and error prone when deriving object speeds on multiple data frames. AM lidar uses a variety of algorithms to calculate object velocity, and the high error rate caused by uneven light and solar glare makes the calculation more complicated.
In contrast, FM Lidar's continuous wave can instantaneously measure speed, reduce computational delay, and consume less power and is easier to miniaturize.
So far, although most of the money has flowed to AM LiDAR companies, Aeva has become a viable competitive technology with its partnership with Audi and Blackmore's strong support from Toyota and BMW. It can be expected that AM and FM will have a battle in the near future.