Now that fully and partially autonomous vehicles are officially a thing in our society, we're testing them to see the results of all the tech, and funny enough, what's happening on city streets with L2 Autonomous (Partial Automation) vehicles is starting to make sense.
Simply put, we react differently when driving an L2 autonomous vehicle, and while it may not "feel" like much, it's enough to raise the number of accidents happening in our cities. New research unveiled by Dr. Francesco N. Biondi of the University of Windsor, Canada, and the Human Systems Lab has shed light on a rather untested subject: our ability to pay attention to actual driving when we're behind the wheel of an L2 Autonomous vehicle.
Overall, a broad range of aspects of driver behavior and function were tested, including "cognitive overload, physiological activation, and visual attention allocation." All this was tracked over the course of 80 minutes during manual (L0) and L2 driving in a 2022 Tesla Model 3 while hauling proverbial butt on a section on Ontario's Highway 401; heart rate variability and steadiness, pupil size and blink rate, and off-road glances were all tracked for this study.
However, if you don't feel like reading the rest of this article since I dive deeper and deeper into how the results are reached, just read the next statement and make your life decisions based on it: During L2 operation, "drivers were more likely to look away from the forward roadway toward the vehicle’s touchscreen. Individual glance durations to the vehicle’s touchscreen were also longer, suggesting that a driver’s visual attention was allocated away from the road for longer."
In short, we're not paying attention to the most crucial aspect of driving: the road and where we're going. "I'm not happy, Bob, not happy."
Photo: Human Factors Assessment of On-road L2 Driving
Now, if you're the inquisitive type and are curious as to how this study panned out and how we got here, welcome; you're my kind of human! In all, 30 drivers signed up to further our understanding of our world and the technology we're bringing to the table.
Since the setting and base tools used in this study have already been mentioned, allow me to point out a bit more about how tracking eye movements and driver attention were carried out: nothing but good old Go Pros (recording at 240 fps), a pair of some very 'funky' glasses, electrodes on a participant's body, some vibrotactile motors, and paired with tons of software to process the details.
The next step was to define all the areas of attention inside the Model 3. Here, four areas of interest were recorded and defined: the side mirrors, rearview mirror, instrument panel, and windshield, the most important section. It was then on to a short training session for participants regarding different functions of the gear they're wearing and functions being tracked, calibrations, and all that, and it was then onto that stretch of highway I mentioned.
Also, allow me to point out that approximately 25,000 vehicles use that predetermined portion of highway on a daily basis. If that's not a real-world test, I don't know what it is. From here, data, data, data, and lots of crunching it! Now, the 37-page paper outlines an insane amount of data, references and cross-references, but to spare you the details, I'll run through some of the more essential details I ran across.
First on the list is subject workload. Once data was processed, findings showed that the mental and physical workload is slightly higher in manual operation (L0) in comparison to L2 driving, showing raised levels of "effort," "physical," and a touch more "mental" strain (check page 23 of the study).
Photo: Human Factors Assessment of On-road L2 Driving
The next piece of data that shows quite the differences is the TEORT, or "Total-eyes-off-road-time." As you can expect, this is a big one and essential to this entire study and anyone operating an L2 vehicle. As opposed to operating an L0 vehicle, L2 drivers take their eyes off the road twice as much. That's insane, especially when we consider that a car accident can happen in a fraction of a second.
Moreover, the distribution of that lost attention goes, guess where? To the dang instrument panel! Maybe you're checking GPS, fiddling with the AC, or searching for that radio station you want. Attention to every other AOI (Area Of Interest), such as side mirrors and rearview, was mostly unchanged, although a tad more time was spent on side mirrors.
Now, Dr. Biondi's details and insights on this study are much more detailed in the full report, but the most important takeaway from this is that L0 operation requires that we take our eyes off the road around 100 seconds, or 4% of the total drive time we engage in. L2 operation doubles that number to 200 seconds over the course of the test, or 8% of total drive time. In short, that's double the chances of you ending up as a statistic.
You don't need to have a Ph.D. to understand the repercussions of this. Also, be sure to check out what brand was reigning supreme in the number of crashes in May 2022, according to the United States NHTSA (National Highway Traffic Safety Administration). I'll give you a hint, Mars.
As for myself, I'll be a tad more aware of what's happening to me when I get behind the wheel of an L2 vehicle, and so should you.