Solving the varied challenges that arise in autonomous driving is an incredibly complex task, but even trying to get started means making sure you have quality, accurate, and well-annotated data. That’s where Scale comes in, having identified early on that the audiovisual industry would require annotation of huge tapes of data, including specialized LiDAR imagery. Today, co-founder and CEO Alex Wang tells me during TC: Mobility 2021 sessions (ExtraCrunch subscription required) that he’s moving to mapping with a new product coming later this month.
“Our role has continued to evolve,” said Wang, of how he works with his transportation industry partners, which include Toyota among many others. “You know, as we work with our clients and we solved an issue for them with labeling data and annotations, you know, it turns out that they come to us with other issues than us. can then help solve also around the data management. , we launched a product called Nucleus for that. A lot of our customers think a lot about mapping and how to deploy with more robust maps. So we’ve built a product, I’ll probably announce it later this month, but we’re helping to resolve this with our customers.
Despite my insistence, Wang did not provide details, but explained in more detail the challenges of mapping and what is missing from the existing maps available to companies working on integrating these with AV systems. which include other signals, such as the fusion of sensors and vehicle-infrastructure components.
“I think a big question for the whole space has been that historically the industry has relied very, very heavily on mapping – we’ve relied very, very heavily on high definition maps. very high quality, ”he said. “The trickiest thing in the world is that sometimes these cards are fake, and how do you deal with that? […] How do you deal with this kind of robustness challenge or updates. Even if you think about it, Google Maps, which is by far the best mapping infrastructure in the world, you know they don’t update quickly enough to [human] Drivers. “
Wang said the challenge is not that different from the one Scale has actively solved for most of its existence, which is that of data stealing. With autonomous driving, it is of the utmost importance to be able to collect and annotate data quickly and accurately, resulting in ever better collection and annotation of future data, and more reliability for assumptions than the system. made on its environment.
“Finding out how to deal with the real-time nature of the changing world is a very important part, a very important part,” he said. While we still have to wait to see what exactly Scale has planned, it seems safe to assume that this is about building confidence in maps and accuracy in mapping as a key ingredient in everything. they launch.