Tag Archives: tesla

Behind the Tesla ‘Full Self Driving’ system

If I were giving advice to somebody considering buying a Tesla at the moment, it would be (a) buy it and (b) don’t believe the ‘full self-driving’ hype… yet.

You’ll be getting a car that is great fun to drive, has amazing range, a splendid safety record, a brilliant charging network, etc… and, in the standard included ‘autopilot’, has a really good cruise control and lane-keeping facility. One thing I’ve noticed when comparing it to the smart cruise control on my previous car, for example, is that it’s much better at handling the situation where somebody overtakes and then pulls into the lane just in front of you. Systems that are primarily concerned with keeping your distance from the car in front have difficult decisions to make at that point: how much and how suddenly should they back off to maintain the preferred gap. The Tesla, in contrast, is constantly tracking all the vehicles around you, and has therefore been following that car and its speed relative to yours for some time, so can react much more smoothly.

The dubiously-named ‘Full Self-Driving’ package is an expensive optional extra which you can buy at the time of purchase or add on later with a couple of clicks in the app. At the moment, it doesn’t give you very much more: the extra functionality (especially outside the US) hasn’t been worth the money. If you purchase it now, you’re primarily buying into the promise of what it will offer in the future, and the hope that this will provide you with significant benefits in the time between now and when you sell the car!

But at sometime in the not-too-distant future, the new version –currently known as the ‘FSD Beta’ — will be released more widely to the general public. ‘Full Self Driving’ will then still be a misnomer, but will be quite a bit closer to the truth. YouTube is awash with videos of the FSD Beta doing some amazing things: people with a 45-minute California commute essentially being driven door-to-door, for example, while just resting their hands lightly on the steering wheel… and also with a few examples of it doing some pretty scary things. It seems clear, though, that it’s improving very fast, and will be genuinely valuable on highways, especially American highways, before too long, but also that it’s likely to be useless on the typical British country road or high street for a very long time!

What Tesla has, to a much greater degree than other companies, is the ability to gather data from its existing vehicles out on the road in order to improve the training of its neural nets. The more cars there are running the software, the better it should become. But the back-at-base process of training the machine learning models on vast amounts of video data (to produce the parameters which are then sent out to all the cars) is computationally very expensive, and the speed of an organisation’s innovation, and how fast it can distribute the results to the world, depends significantly on how fast it can do this.

Last week, Tesla held their ‘AI Day’, where Elon Musk got up on stage and, in his usual way, mumbled a few disjointed sentences. Did nobody ever tell the man that it’s worth actually preparing before you get up on a stage, especially the world stage?

However, between these slightly embarrassing moments are some amazing talks by the Tesla team, going into enormous detail about how they architect their neural nets, the challenges of the driving task, the incredible chips they are building and rolling out to build what may be the fastest ML-training installation in the world, and the systems they’re building around all this new stuff.

For most people, this will be too much technical detail and will make little sense. For those with a smattering of knowledge about machine learning, you can sit back and enjoy the ride. There are lots of pictures and video clips amidst the details! And for those with a deeper interest in AI/ML systems, I would say this is well-worth watching.

There are two key things that struck me during the talks.

First, as my friend Pilgrim pointed out, it’s amazing how open they’re being. Perhaps, he suggested, they can safely assume that the competition is so far behind that they’re not a threat!

Secondly, it suddenly occurred to me — half way through the discussions of petaflop-speed calculations — that I was watching a video from a motor manufacturer! An automobile company! If you’re considering buying a Tesla, this is a part of what you’re buying into, and it’s astonishingly different from anything you’d ever see from any other car-maker. Full self-driving is a very difficult problem. But this kind of thing goes a long way to convincing me that if anybody is going to get there, it will be Tesla.

You may or may not ever pay for the full FSD package, but it’s safe to assume much of the output of these endeavours will be incorporated into other parts of the system. So, at the very least, you should eventually get one hell of a cruise control!

The livestream is here, and the interesting stuff actually starts about 46 minutes in.

Altruistic Autonomous Vehicles

One of my shortest recent posts generated quite a lot of discussion, both here and on Facebook. I wrote:

When we have proper and affordable self-driving vehicles, will that be the end of the railways?

Clearly there are some things that railways will do better for the foreseeable future, like long-range high-speed links, or carrying heavy freight. And don’t get me wrong: I like train journeys. But it seemed to me that the key reasons people currently take trains for normal day-to-day journeys — wanting to read en route, a lack of parking at their destination, avoiding congestion — could very soon be overcome when, for example, your car can go and valet-park itself after dropping you off at the office.

And the disadvantages of train travel: the fact that instead of going from point A to point B, you have to go at least from point A to point B to point C to point D, possibly waiting on a cold platform at point B for an indeterminate period, and not being sure whether you’ll get a seat from point C to B on the return journey. Will it be worth the hassle?

One of my assumptions is that traffic congestion will become less of an issue when cars are smarter, of course, which may not be a valid one, especially if lots of train travellers take to their cars instead.

There’s an interesting question as to whether lots of small independent agents trying to meet their own goals are going to result in an optimal solution for road congestion as a whole. We may start off with vehicles that are pretty autonomous initially, but become less so in due course, as the road infrastructure starts to adapt to them. Network packets on the internet know their destination, but it’s the routers (the junctions) that tell them which exit from the roundabout to take.

Will the Department of Transport manage overall use of the network better than each individual car? Well, that depends on who has the better computer scientists, of course! But it also depends on the amount of knowledge each vehicle can get about the overall road network, and, of course, on how selfish your car is: will it decide that the needs of the many outweigh the needs of the one (or the few)?

Perhaps we need Altruistic Autonomous Vehicles? (You heard it here first!) There could be financial incentives to encourage this. You get lower road tax if your car agrees to obey centrally-prescribed rules at times of high congestion. Perhaps you get to use the high-speed autonomous-only lanes if you’re willing to hand over to the cloud-based algorithms. Of course, this could open up all sorts of wonderful opportunities for hackers, too. Remember the movie?

Anyway, perhaps congestion will be less of an issue, but for a completely different reason. If you can be having a coffee, working on your laptop and taking Skype calls while you slip quietly along in your electric car, you may be more productive than if you’d got to the office on time.

The new Duracell

I had the honour of meeting Elon Musk briefly some years ago. Back then, he was just doing space exploration. There aren’t very many people for whom you could write that sentence – even without the word ‘just’ – but for him, its inclusion is entirely appropriate.

Now, he seems to be doing… well… everything that’s cool. I have since admired his orbital exploits, ridden in his cars, and, of course, bought a significant proportion of my purchases using Paypal – which helps pay for all the rest.

The latest product is apparently to be batteries – here’s the new press announcement – but these are not your average AAs. They’re wall-mounting.


At 10kWh, these could run a lot of LED lightbulbs for a long time. Especially if you’re not also using them to recharge your model S.

They’re designed to make it easier for solar-equipped households to depend less on the grid, especially by time-shifting the peak sunshine energy at noon to the morning and evening, when there’s peak demand. That’s not simply about being green: just a few weeks ago, visiting friends were telling us about the ‘loadshedding‘ powercuts in Cape Town recently, where everybody got a scheduled two-hour outage each day to help cope with the insufficient power-generation capabilities in the country. This seems perfect for that, too.

Still, the next challenge Musk has to address is the really tricky one: the manufacture of solar panels which don’t ruin the appearance of the building to which they’re attached.

© Copyright Quentin Stafford-Fraser