During Reuters Events Automotive USA 2023, AutoVision News Radio host Carl Anthony went on a ride-and-drive demonstration with LeddarTech. While navigating downtown Detroit, LeddarTech engineers showcased the company's perception platform, specifically the low-level sensor fusion architecture.
Following the demonstration drive, Carl talked with Pierre Olivier, LeddarTech's Chief Technology Officer. During the discussion, Pierre shared more about LeddarTech's focus and how the company identified certain key market trends. Pierre also helps Carl further understand how low-level sensor fusion works and shares the legacy he and LeddarTech wish to leave for the future of mobility.
LeddarVision Sensor Fusion E-Book: https://tinyurl.com/2ve5dh86
Follow AutoVision News on LinkedIn: https://tinyurl.com/49jyrd3b
[00:00:00] My name is Carl Anthony, and I work in the automotive industry in Detroit.
[00:00:06] Sometimes that work encompasses future vehicle technology,
[00:00:10] and that's what we talk about here, for the most part anyway.
[00:00:14] This is AutoVision News Radio.
[00:00:17] As part of Reuters' events, AutomotiveUSA 2023,
[00:00:23] Lettertech hosted a ride-and-drive demonstration for attendees. after speaker and panelist, Pierre is often featured on the agendas of automotive technology conferences throughout North America and Europe. I'm pleased to share my interview now with Pierre during Reuters events, automotive USA 2023. Moving at the speed of mobility, this is AutoVision News Radio with Karl Anzani in Detroit, Michigan.
[00:01:42] So that our tech, well actually first of all,
[00:01:44] we've been in business robotaxis with 20-30 sensors around the vehicle and these are very, very costly solutions both from sensor set standpoint but also from all How did you see that trend here? What was it in the market that led you to make that decision and go that way as a company? That's a great question. And actually if you go back a couple years ago, certainly many of the OEMs were working on the assumption that they would be able to start from level 5, level 4, the Rob for example, AUD type applications, but we also have four cameras that cover the complete 360 degrees around the vehicle. So two on the mirrors and then one in way to achieve it. And that's where we're able with limited number of sensors to demonstrate the level of performance that you've seen on the road today. To that point, we did take a trip around downtown Detroit, outside of the Huntington Place,
[00:07:02] went past some Detroit landmarks
[00:07:04] like Ford Field and Tiger Stadium.
[00:07:06] We were inside the vehicle you'd want to break if they started
[00:08:22] accelerating.
[00:08:23] So these are the types of objects that we provide to customers. classical way of doing sensor fusion is, which normally each sensor would have its own built-in smarts. So for example, each radar would extract object detections and would send those over, for example, the canvas. And then they would be combined by some central computing.
[00:09:41] But the issue with doing that is as each taking these individual radar reflections, and we're merging them into a combined image. So this image has both the 2D data, of course, on the cameras, but also the 3D data from the radar,
[00:11:02] or it could be LiDAR in a higher level of automation.
[00:11:04] So that's what we it requires, for example, of course, the different sensors to be very accurately calibrated with one another. So some of the things we did, of course we developed and we patented some of these
[00:12:22] base algorithms, but also, what is on the horizon?
[00:13:40] How should I say what goals do you have going forward maybe in 2024 and beyond? legacy mobility into this new world of basically addressing not only the safety challenges but also the different societal challenges how you get like for example elderly people to be mobile how you get the people in cities to be more mobile and so on so really what what I'd like to be able to say is we I've
[00:15:03] contributed and we've contributed to making transportation better for and