Tesla versus Audi autopilots

I am in love with the idea of the active suspension in the new Audi A8, but maybe the autopilot is the more significant innovation?

“Audi Beats Tesla (And GM) To Level 3 Autonomy” (Seeking Alpha):

Audi’s forthcoming version of the A8 sedan is the first production automobile to arguably achieve Level 3 autonomy. … The key point in moving from Level 2 to Level 3 autonomy is that Level 3 expects that the user only has to intervene whenever the car is not able to handle a situation and asks for the user to take over. The vehicle for its part must be able to perform all activities related to driving the vehicle at other times.

Now to be clear, Audi’s solution has certain limitations. For one it is only meant to be activated in relatively slow moving traffic (up to 60 km/hr) and only on divided roads where a physical barrier separates vehicles driving in the opposite direction.

The new Audi A8 will be the first production vehicle to use LIDAR along with cameras, radar and ultrasonic sensors.

Audi’s solution is the first of its kind to require absolutely no monitoring of the vehicle as long as the driver is available to take over when the vehicle asks him/her to. While the vehicle is driving itself, the customer is free to do anything else, for example even watch a movie on the on-board TV screen.

Tesla’s Autopilot also requires the user to keep their hands on the steering wheel and asks the user to monitor the actions of the car. This makes it a SAE Level 2 solution at best.

I’m wondering if an autopilot that works from 0-35 mph is useful. On a recent trip to Dallas/Ft. Worth we experienced highways clogged at all hours on both Friday and Saturday. After returning to Boston we found the central highway network jammed up and sat for an extra 30 minutes in stop-and-go traffic on the Mass Turnpike (I-90). However, traffic did occasionally speed up to faster than 35 mph in both cities and therefore we would have been turning the Audi autopilot on and off every 5-10 minutes.

Readers: What do you think? Is this Audi feature likely to be useful in practice?

[Separately, maybe Audi should be donating money to pro-immigration American politicians. If via immigration and the children of immigrants the population grows to 450 million, Americans will spend nearly all of their highway driving time below 35 mph. (Of course we have a lot of open space in the U.S., but immigrants, like native-born Americans, tend to cluster in cities.) See also How much would an immigrant have to earn to defray the cost of added infrastructure?]

8 thoughts on “Tesla versus Audi autopilots

  1. Ironically, the auto-pilot features of my Honda (“level 2” at best) are most useful above speeds of 60 km/h. While the distance tracking still works reasonably well below those speeds, the lane tracking never does.

  2. If you’re doing the 93 commute, you’ll spend a lot of time in this speed range, probably 30-60min a day. I’ve found myself wanting a feature like this before. The Delphi radar-based adaptive cruise solution is available on many cars now, and works at this speed range, so could be an alternative. It’s not full autonomous obviously, so you’d still have to steer, but in traffic steering is much less tiring than the control loop involved with accelerating/decelerating and maintaining a variable speed.

  3. “Audi’s solution is the first of its kind to require absolutely no monitoring of the vehicle as long as the driver is available to take over when the vehicle asks him/her to.”

    The statement above is self-contradictory.

    I rather agree with the following:

    “Sometimes Teslas are dangerous because of what they call ‘autopilot,'” says Wozniak. “You get thinking, ‘Oh, it is easy, I can reach over and not look for a few seconds,’ and that is the second your car drifts over the line,” he says, adding that it is “easy to make mistakes, especially certain weather conditions and whatnot.”

    and:
    “Driving my Tesla, over and over and over there are unusual situations on any road anywhere and every single human being alive — dumb or smart — would be able to get through it and the Tesla can’t,”

    Perhaps, Audi’s autopilot is better, but I doubt it — the technology is just too immature and unreliable to “require absolutely no monitoring”.

  4. And yet tesla confidently sells full self capability option right now (try to configure model suitable or x)

  5. The Eyesite feature in my 2015 Subaru Forester will not steer, but the adaptive cruise feature of this system will maintain a safe distance with the car in front from at least 75 mph (highest speed I have ever set it for) right down to 0 mph. When the car in front decelerates to a temporary stop for approximately less than 2 seconds, the cruise will remain engaged and automatically start pacing the car ahead when traffic starts moving. Subaru does not describe this behavior in the user manual. But, It is sure handy to engage it in stop and go traffic when you are tired or want that extra margin of safety.

  6. The Audi system is pretty good, and I had pretty good luck with the more basic features in my 2015 model. Adding LIDAR is going to be an advantage in making these systems work.

    The innovation with a larger impact is the use of automatic emergency braking. We need this soon in all cars. I regularly travel on interstates where the traffic goes from 50 to 0 with little warning. With everyone staring down at their phones, the number of accidents is increasing. I was hit 2 months ago… before everyone gets this though, we’re still at risk from the cars that don’t have it. I am thinking about adding some kind of device that warns the car behind me that it is approaching too fast based on my deceleration. Maybe a light and sound, or just a rapidly inflating air bag on the back end… https://jalopnik.com/this-concept-car-has-airbags-on-the-outside-1819853012

Comments are closed.