Tesla Versus Porsche Duel Misses The Boat; Focus Should Be Self-Driving Capability

It’s the car duel of the century or the skirmish at the Nurburgring racetrack, the seemingly breathless competition that’s grabbed the attention of car aficionados, pitting the newly unveiled Porsche Taycan, an EV sports car, against a Tesla.

Here’s the sordid tale.

Porsche unveiled its high-performance EV in early September and showcased the Taycan’s sportiness by providing the car’s performance times based on racing around a legendary track, Germany’s Nurburgring, an especially demanding course that’s a purist’s dream for both professional and amateur car racing.

Elon Musk then opted to prove Tesla’s mettle by having a modified Tesla Model S rocket around the Nurburgring racetrack, and eyewitnesses claim the Model S readily bested the Taycan’s recorded times by perhaps 20 seconds.

Right away, some said the Tesla Model S breakneck time shouldn’t yet count since it was not an “official” recording of trial time and thus it was merely based on (biased) eyewitness conjecture.

Furthermore, many have pointed out that the Model S had been modified for racetrack running and thus served as a contrived and outright unfair comparison to the factory-version Taycan. The assertion is that an everyday off-the-production-line Model S should be used rather than a souped-up version.

The tussle continued when one of the Teslas being prepared for the racetrack appeared to break down mid-track and sat halted on the racetrack. Meanwhile, by seeming coincidence, a Porsche Taycan drove past it and a picture worth a thousand words was snapped. The pic has become the talk of the town in that car buffs who are against Tesla point out the superiority of the Taycan zipping past a dead-in-the-water Model S.

It’s an easy bet that this macho drum-beating affair is going to continue. “My car is faster than your car” is the mantra of those who cherish their sporty vehicles.

One aspect to keep in mind and that often gets scant attention is that these racetrack times are not solely reflective of the car itself since it takes a human driver at the wheel to produce those racetrack shattering performances.

The person sitting in the driver’s seat will make a significant difference in terms of how fast or how slow these sporty cars complete the course. I mention the role of the human driver because this sparks another point that might be somewhat unspoken and hidden from view.

Here’s the gist: Should we be preoccupied with raw speed or would it be more fruitful to conduct car competitions focused on self-driving tech and how well one car compares to another as a self-driving car?

In other words, maybe we are missing the boat by continuing to be excited about how fast cars can go, and instead, we ought to shift that enthusiasm and attention toward the future, namely the advent of true self-driving cars.

Let’s unpack the matter.

Caring About Self-Driving Cars

Before launching into a heated debate about how to best compare tomorrow’s state-of-the-art cars, it is important to clarify what I mean when using the phrase “self-driving cars” throughout this discussion.

There are semi-autonomous cars that require a human driver to be present and able to drive the car, doing so in a co-sharing arrangement with the automation of the car. These are often referred to as Level 2 and Level 3 cars and make use of ADAS (Advanced Driver-Assistance Systems).

Even more cutting-edge are the truly autonomous cars, ones that are self-driving and there is no human involved in the driving task. These driverless cars are considered at the Level 4 and Level 5 of self-driving autonomy (none fully exist as yet, and the roadway trials taking place are at best in the lower end of a Level 4).

A true self-driving car is one that meets the requirements of a truly autonomous car. Rather than repeatedly having to say the word “true” in front of the phrase self-driving car, just assume that a self-driving car is one that truly is autonomous.

For the semi-autonomous cars, the variety of ADAS features is quite expansive and there are essentially few if any standards governing how they work and what they do.

As such, trying to compare a car by automaker X that has its own set of ADAS is often hard to do in contrast to the ADAS of a car made by automaker Y. The differences are apt to be voluminous, plus each feature will have some positives and some negatives, making it problematic to say which is the better of the comparison.

I’m going to focus therefore on the notion of comparing self-driving cars rather than trying to compare semi-autonomous cars.

Your first thought might be that if a self-driving car is autonomous, shouldn’t any and all self-driving cars simply drive in the same ways.

Sorry, that’s not going to be the case.

Just as human drivers can drive cars in differing ways, so too will the AI driving systems be driving self-driving cars in differing ways.

Each automaker and tech firm that is developing a self-driving car are doing so in their own proprietary manner. This means that the driving style and approach for the various brands of self-driving cars will differ. In addition, it is likely that even the various models of a brand will be driving in differing ways.

I realize that this might seem like shocking news.

Many in the media seem to be suggesting that all AI driving systems are essentially cloned robots and will be driving in the same precise manner.

Not true.

As such, we now have a basis to consider an intriguing idea, perhaps we ought to be getting enamored about how AI driving systems are doing and showcase their “talents” by having them compete against each other, similar in a manner akin to comparing human-driven cars.

Competing Self-Driving Cars

The rapt attention that goes toward how fast cars can go is a business driver, one might say, pushing the automakers towards making faster and faster cars.

Yes, it’s fun to talk about the speed that a car can take a given curve or zip down a straightway. There are bragging rights to be had. Buyers of cars are proud of their possibility of outgunning other cars on the roadway.

Of course, the reality is that most of the time you really are unable to fully exercise all that power under the hood of your car. Being stuck in commuter traffic does not especially highlight how fast your car can go and nor how quickly it can merge with other stop-and-go traffic.

You are sitting on top of an unleashed tiger, which may from time-to-time you get a chance to let loose, but most of the time it is caged and untapped.

Perhaps we should consider another direction rather than speed to push the automakers and tech firms toward.

Suppose we all got excited about the AI driving capabilities of self-driving cars.

Imagine if the news was filled with the latest breakthroughs of how a self-driving car navigated a dicey driving situation in a crowded downtown area. Or, how deftly a self-driving car managed to drive in a school drop-off zone and weaved safely around the crazy parents dropping off their children.

Instead of putting cars onto racetracks, perhaps we ought to be putting self-driving cars into proving grounds or closed tracks that are intended to test out AI driving capabilities, and then be excited about how those AI systems are doing there.

Indeed, the automakers and tech firms are already making use of proving grounds, yet it is something being done in a quiet manner and without any fanfare. Sure, it makes sense to be out of view when doing testing, and so it is certainly understandable that we don’t hear much about it.

Once the self-driving car capabilities are mature enough, sufficient to be putting them onto our public roadways (which is a controversial act and some believe we are all guinea pigs in a dangerous ongoing public experiment), maybe we ought to become excited about comparing those self-driving cars on their closed track performances.

Metrics For Self-Driving Car Performance

It would be unseemly to focus on comparing self-driving cars solely by how fast they go, and thus we’ll need to consider other factors of performance.

I’m not suggesting that the AI shouldn’t ultimately be able to drive fast, which is fine when appropriately undertaken.

Let’s though first aim at driving legally and safely, and we can then come back around to the speed factor later.

We’ll need to be interested in the more mundane acts of driving that we all take for granted.

Everyday driving such as:

·        How soon did a self-driving car detect and avoid a pedestrian that was jaywalking?

·        Did the self-driving car merge with human-driven cars in a manner that kept the traffic flowing?

·        Was the AI able to drive in a cramped parking lot that had lots of cars and people, and did it do so without bumping into anyone or anything?

·        Etc.

One of the issues facing the advent of self-driving cars is that there are not yet any accepted standards about what constitutes a safe and road-ready self-driving car.

Currently, the automakers and tech firms that are building self-driving cars can pretty much decide whether they believe that their driverless cars are road-ready or not.

Most of the time, a human back-up driver is included in the self-driving car as it is undertaking public roadway tryouts, though having a human back-up driver does not guarantee the avoidance of crashes (as per the famous Uber incident in Arizona).

Thus, it is in the eye of the beholder as to whether a self-driving car is ready for prime time.

Some believe that this is too loosey-goosey and we are going to regret not having established a kind of Turing Test for self-driving cars (for my article on this topic, use this link).

Conclusion

Overall, the point herein is that instead of being overly preoccupied with the pursuit of fast rides, it might be handy and helpful to refocus some of our energies toward self-driving cars and a future of automated driving.

This reminds me of the famous joke about the man in a parking lot at nighttime that is furtively looking for his lost car keys, doing so under the light emanating from an adjacent light post.

Another person comes along and offers to help look for the keys.

After a few moments of combined searching, the helper asks where the keys were approximately dropped. The man points toward a distant and darkened area of the parking lot and says that he dropped the keys over there.

Perplexed that they aren’t searching in the darkened area, the helper asks why then they are searching in a spot that is unlikely to be hiding the keys. The man brashly offers that it’s because this is where the light is.

Drum roll, please.

Perhaps we’re right now looking at the speed of cars because that’s where the light happens to be shining, but we ought to be looking where the keys really are, namely residing in the advent of self-driving cars.

Don’t interpret this proposition to mean that we can’t still have fun about human-driven cars and how fast they can go.

Presumably, even once we have widespread AI-driven cars, the odds are that we humans will still be fascinated by the incredibly fast driving of human-driven cars.

Far off in the future, it is forecast by some pundits that we won’t have any human-driven cars at all. In which case, maybe we’ll be going to racetracks to watch AI systems driving self-driving cars at breathtaking speeds, rather than seeing human drivers do so.

Or, we might have racetracks that specialize in the old-time human-driven approach and for the sake of tradition allow humans to drive sporty cars. The populous at large might have decided that human drivers should never be on public streets, but if human drivers want to drive around a “silly” racetrack, so be it.

There are some that believe we’ll never reach a time when we’ll ban all human driving from our roadways and that a mix of AI-driven cars and human-driven cars will always be allowed.

For now, we’re bound to continue to compare racetrack times of one car versus another, and perhaps we’ll gradually get equally excited about how one brand of self-driving car did a greater job at everyday driving than did a competing brand.

I’ll make a champagne toast to that moonshot race-winning feat for self-driving cars.