Safety Through Synchronization ft. Paul McGlone of Seeing Machines

[00:00:00] My name is Carl Anthony, and I work in the automotive industry in Detroit.

[00:00:07] Sometimes that work encompasses future vehicle technology, and that's what we talk about

[00:00:12] here, for the most part anyway.

[00:00:14] This is AutoVision News Radio.

[00:00:19] On February 2, 1993, my parents gifted me a blue, Webster's concise English dictionary

[00:00:26] from Barnes & Noble. Common factors in aerodynamics is the application of psychological and physiological principles to the engineering and design of products, processes, and systems. So how do we take what these definitions are saying and apply them to the vehicle and apply them in a way so that everybody who is inside that vehicle gets home safe?

[00:01:43] This is the core question at seeing machines CEO Paul McLone.

[00:03:04] As a leader of one of the Australian capital territories, for engineering science. Those two founders are still with us today, which is amazing. The company focus is core area of endeavour around eye tracking and in particular the relationship between humans and machines. There's human machine interface. Today, we're focused on transport. We operate in three vertical markets within all segments within transport. That's

[00:05:26] Why is it important that when we talk about ADAS, we also make room in the conversation for something like DMS?

[00:05:28] Yeah, that's a great question.

[00:05:31] It's one that's been asked of us more and more frequently by our customers.

[00:05:36] I put this in the realm that ADAS is still a really important area and probably the acronym

[00:05:43] that gets most talked about certainly more than D of the conversation today. And in fact, a reality in new cars. The first application for seeing machines was in the mining industry, where vehicle operators are driving monotonous routes for hours on end, leading to fatigue and distraction.

[00:07:00] And mining indeed was an excellent proving ground for seeing machines given its importance In this environment, we've got a pretty advanced safety system, of course. It's scanning the exterior of the vehicle, sensing imminent threats, and is able to make real-time decisions. But there's a big assumption around taking back control, or what the driver is doing in order to be alert to the problem, and ultimately prevent or minimize accidents.

[00:08:22] In these kinds of systems, the main role is to, of action, specifically if it can be done in a way that avoids the vehicle making its own decision, either by, you know, applying braking or what have you. And if we consider airbags, as I just mentioned, airbags can be adjusted and deployed using signals from the seat sensor and the seat belt sensors.

[00:09:42] But imagine an environment where the vehicle understands the size of the driver or the And it's almost unending. First and foremost, there is a requirement for a deep understanding of how humans act and interact and react in a naturalistic environment, like a driving environment. So from our perspective, we've got a deeply experienced team of scientists, human factor scientists

[00:11:00] who specialize in the HMI human machine interface design.

[00:11:06] Some important elements of this technology. A lot of words, but human factors design is mission critical as a starting point. Now, when you apply this understanding, that's knowing the driver or the occupant, knowing what they're doing, what they're doing within a context, where they are, what their attention state is at any given point of time.

[00:12:22] Well, then there's an opportunity to feed those signals or of this whole sensing integration opportunity. As Paul mentioned, seeing machines has a team of human factor scientists who apply this understanding to the machine interface. To date, seeing machines has acquired over 14 billion kilometers of naturalistic driving data, including more than 40 studies across simulators, test tracks, and on-road environments.

[00:13:43] These data sets inform seeing machines as they develop optics and processing technologies enhance the user experience. Because we know that we can develop technologies, frankly, relatively easily, do certain things, observe certain things. And if we're not concerned about the user experience, well then, you know, we can certainly drive a whole range of outcomes

[00:15:01] that would be positive. The issue is humans help to better protect passengers, pedestrians, and cyclists across the EU, expectedly saving sit on the advisory board for this protocol. So, we've worked for several years now very closely with EuroInCAP. In fact, we prevented the first science to EuroInCAP and European Commission many years ago,

[00:17:44] and the objective there was to demonstrate that there was a way to intervene very, very long road map that will be driving safety in this space. And, yeah, we're really pleased about that. I go back and flip through my Webster's concise English dictionary, the gift from my parents in 1993 that I've had for all these years. Page 384, the word purpose is right above the word per.

[00:19:01] Of course, there's a cat example, but another supplemental example given for per is a low

[00:19:06] noise made by a powerful engine. that was your baby from the very first thought that everyone may have had in that laboratory of what seeing machines could become with that purpose all the way up to today where they're helping shape legislation and over 1.3 million cars feature seeing machines technology. What I wanted to know from Paul is how do you do it?

[00:20:22] How do you make it work? To be financially viable, we have to develop products that are commercially saleable. We break down this notion of being sustainable into, let's say, half a dozen more granular metrics that are linked to financial performance, and then suddenly you've got a brilliant way to link purpose to what somebody does.

[00:21:40] And we do that across each of the five or six categories that I mentioned.

[00:21:43] Now that's a formal process that we engage in here.