Examining How Self-Driving Cars Will Drive Up EV Adoption

AUG 11 2018 BY VANJA KLJAIC 17

Everything points to electric self-driving cars

It was back in 1997 when the Toyota Prius was released as the world’s first mass-produced hybrid. Almost 15 years later, the Tesla Model S followed. But, unlike the Prius, the Model S brought on something completely new to the market: autonomous driving. Certainly, it wasn’t available from the start. Nor is it fully autonomous. But, currently, the autonomous driving in the Model S – unlike some other, self-driving autonomous cars from other car makers – actually works rather well, in most situations.

But, for self-driving cars to accelerate EV adoption rates, we’ll need to see a collective effort from everyone involved. Policymakers will have to establish new rules and regulations, gunning for even cleaner vehicles. Car makers and tech companies will need to raise the bar in both battery technology and programming. And lastly, consumers will need to buy into an idea that clean, self-driving vehicles are easier to use, less expensive to operate and, in the long run, will benefit everyone. Naturally, there are several benefits of self-driving cars, as pointed out by the Green Car Congress.

The world-renowned green tech transportation website gives a few fairly reasonable guesses. They cover the pros of today’s electric vehicles, giving us an insight that bases its validity on reduced maintenance costs, lower fuel expense and – most importantly – less pollution. Today’s electric vehicles – even with its level of autonomous driving – are still far better at the wheel than an average human being. This allows for fewer accidents. The cars utilize an array of sensors, radars, and cameras to virtually scan the road. Then, by utilizing the impressive processing power under the hood to handle the data streams from all those input sensors allow the vehicle to react faster in various emergency situations. In turn, this makes autonomous vehicles a lot safer to operate for both the driver and the passengers.

Additional arguments towards self-driving vehicles pushing EV adoption rates, discuss several interesting prospects. The sheer thought of gas-powered, self-driving vehicles turns into a dystopian thought about the future. After all, some experts feel that such vehicles could wreak havoc on the environment, causing a 200 percent increase in emissions. Battery power and renewable energy production is the key to sustainability and less impact on the environment.

The article covers a lot of particular topics. It makes for a few fairly argumentative approaches to well-known problems. If you’re into problem-solving and electric cars interest you, then this article is definitely a good read.

Categories: General

Tags: , , , ,

Leave a Reply

17 Comments on "Examining How Self-Driving Cars Will Drive Up EV Adoption"

newest oldest most voted
Pushmi-Pullyu

“…the autonomous driving in the Model S – unlike some other, self-driving autonomous cars from other car makers – actually works rather well, in most situations.”

The only situation in which the cars Tesla sells to the public can perform anything near “autonomous driving” is after the human driver has driven it onto a freeway, and even then only so long as he doesn’t need to change lanes, or exit, or choose a path at a “Y” intersection, or encounter any situation where confusing lane markings might lead the car to hit a solid obstacle… or even encounter parked vehicles and other stationary obstacles in the lane in which it’s driving.

I think most people would call that still pretty far away from “autonomous”.

* * * * *

Tesla Autopilot is always “on” whether it’s controlling the car or not. What the driver can choose is whether or not AutoSteer is activated. When it is:

Can Teslae (Tesla cars) change lanes without being told to do so by a human driver? No.

Can Teslae obey stop signs and stop lights? No.

(continued…)

Pushmi-Pullyu

(…continued from above)

Can Teslae follow driving instructions from Google Maps, or any other in-car navigation system? No.

Can Teslae take active measures to avoid an accident? Well yes; but other than slowing down and/ or moving from side to side within a traffic lane… No.

* * * * *

Teslae have pretty good Level 2 semi-autonomous driving systems, with perhaps some limited aspects of Level 3.

But to be truly autonomous, cars need reliable Level 4/5 autonomy. We’re at least a few years away from that, and likely at least several years. And it looks to me like Waymo/Google has more advanced semi-autonomous driving tech than Tesla does. I absolutely would not bet that Tesla will be the first to achieve reasonably reliable* Level 4 autonomy.

*By “reasonably reliable” I don’t mean 100% reliable; that’s an impossible goal. I mean “reliable” in the sense of having an overall accident rate somewhat lower than the average human driver.

antrik

Stop splitting posts. It’s annoying.

FWIW, one of your points is supposed to change really soon now: changing lanes without human input.

In general, it’s very hard to compare Tesla’s progress with Waymo’s or Cruize’s, since they are taking completely different approaches. Tesla has limited functionality, but fundamentally capable of working everywhere; while Cruize has complete autonomous functionality, but reliable only in controlled environments. Tesla gradually adds more functionality; while Cruize gradually adds broader applicability.

Kacey Green

PMPU I’d at least give them until next month’s rollout of version 9 and the first of the AP2 features to see how far out they are, yes all that stuff is true today but they really just have three implementations of AP1 on three different publicly released platforms plus the prototypes of AP3

Pushmi-Pullyu

It’s been quite a few months since Tesla released a video showing one of their advanced test platform cars with a higher development of AutoSteer; one capable of recognizing and obeying both stop lights and stop signs, and also apparently capable of navigating, taking turns to get to a destination.

But even that video showed a shocking number of “false positives” from the optical object recognition system. Literally hundreds of trees at the side of the road were falsely identified as being “in-path” objects! I’m a strong fan of Tesla, but it looks to me like their quest for autonomous driving is mostly stalled out. (But I’d love to be proven wrong!)

I wish that various auto makers including Tesla would pool their efforts, and develop a fully autonomous driving system that will work on all street-legal cars; not individual different systems from different auto makers. Having a bunch of different companies developing autonomous driving independently means there will be a lack of uniformity in the behavior of self-driving cars. That’s a pretty stupid approach if the primary goal is to reduce accidents. How self-driving cars work should be very predictable to other self-driving cars. They should all work the same way.

Roy_H

“I wish that various auto makers including Tesla would pool their efforts, and develop a fully autonomous driving system that will work on all street-legal cars; not individual different systems from different auto makers. Having a bunch of different companies developing autonomous driving independently means there will be a lack of uniformity in the behavior of self-driving cars. That’s a pretty stupid approach if the primary goal is to reduce accidents. ”

I disagree with this portion. Competition is a great way to spur development. After a few years, the best methods will be adopted by the other manufacturers and your goal of similar behavior will be realized. Ongoing advancements will create small differences so total uniformity will not exist.

SJC

Competition is a way to kill your competitors, it brings duplication, patents, legal actions and waste.

antrik

I agree that it would be better for everyone if all car makers pooled their development resources in a Free (Open Source) Software project… Unfortunately, commercial players are usually unwilling to do so with anything they consider a core technology, since the large players are hoping to gain a competitive advantage; while the smaller ones who can’t afford that, just outsource development to suppliers entirely…

jim stack

as has been reported here Waymo bought 120,000 Chrysler Pacifica Plugin hybrid vans and 20,000 Jaguar I-Pace 100% electric vehicles for self driving. The GM owned drive automation uses all Chevy Volt and Chevy SPARK EV’s for their Self driving. Uber uses the Volvo Plugin hybrids. Plugging in along with Self Driving is the Future.

Pushmi-Pullyu

I’m eagerly awaiting news of what happens with Waymo’s pilot project for autonomous driven cars; a fleet of self-driving taxis in a Phoenix area suburb. That’s one of the things that makes it look to me like Waymo is significantly ahead of Tesla in development of autonomous vehicles.

As I’ve said repeatedly, I think Tesla is going to be stuck at its current stage of development so long as they refuse to put either lidar or a high-res radar array into their cars. And over the course of several months now, I haven’t seen anything suggesting I’m wrong.

Roy_H

I agree that the existing system seems inadequate, but as you know I believe binocular vision using triangulation to determine distance to objects would be a superior option. Cameras have higher resolution than radar or lidar.

antrik

See my other post: neither is clearly ahead — they are just approaching it from completely different angles.

You could just as well claim that Waymo seems stuck, since they had fully autonomous test fleets on the road like five years ago, yet still only planning deployment in a controlled environment for now…

Let me remind you of Elon’s argument: lidar seems easier in the start, since it allows getting a limited view of the surroundings without very complex processing. However, this is not sufficient for full autonomy outside of very well-known environments. For generic self-driving, you still need to get powerful passive vision. Since Tesla is relying on passive vision from the start, they don’t get an early lead — but that doesn’t necessarily mean they will be late to the finish line…

Doggydogworld

Elon’s argument is BS. He can’t put ugly, expensive LIDAR on his premium sedans, so he sells what he has.

What’s so limited about the Waymo/Cruise environments? It’s not like they have to redesign the system to move from Phoenix to Nashville..

antrik

The point is that every time they want to open a new market, they have to precisely map out the entire area (especially intersections etc.), and test all possible scenarios everywhere in this area. They can’t just decide to go nation-wide the next day. The technology is just not there yet to work reliably in unknown areas.

Dav8or

Uh-oh, sounds like you’re betting against Elon Musk. You know what they say…

SJC

AV is not what is holding up EV adoption.

antrik

Of course not. But it should encourage it.