Tesla Claims Are Out Of Sync With Reality

| About: Tesla Motors (TSLA)

Summary

Automatic lane change is dangerous and probably illegal without rear radar.

According to a leading autonomous vehicle engineer, no radar currently in the market is useful for automatic lane change.

The company has a history of announcing and then forgetting major features and projects.

First things first: Tesla (NASDAQ:TSLA) has claimed the Model S, when equipped with Autopilot hardware and the concomitant software, can change lanes just by tapping the turn signal - i.e., no steering required on the driver side. From the official blog post:

Model S will be able to steer to stay within a lane, change lanes with the simple tap of a turn signal, and manage speed by reading road signs and using active, traffic aware cruise control. It will take several months for all Autopilot features to be completed and uploaded to the cars.

Notice the will, clearly indicating that these are not just possible or potential or likely features, but things the company has carefully planned for. Notice also the months, which any reasonable person would interpret as less than a year (i.e., before October 2015).

And what is this Autopilot hardware?

The launch of Dual Motor Model S coincides with the introduction of a standard hardware package that will enable autopilot functionality. Every single Model S now rolling out of the factory includes a forward radar, 12 long range ultrasonic sensors positioned to sense 16 feet around the car in every direction at all speeds, a forward looking camera, and a high precision, digitally controlled electric assist braking system.

Notice the forward before radar. There is no rear radar, and sonar can only see 16 feet around - so how is the car supposed to check if there are cars coming in the adjacent lane?

Think about it. If I'm driving at 50mph and on the left lane, there is somebody coming in at 80, he's reducing the distance between us by 30mph - or 44 feet/second. In other words, sonar would be able to warn you about the incoming car 0.36 seconds before collision. Let's not even talk about the Autobahn.

(And before anybody brings up the backup camera, all Tesla statements seem to indicate it is not connected to the Autopilot system in any way - and besides, cameras are much worse than radars for judging distance).

I have emailed Tesla asking if a rear radar is in the offing. I'll update the post if and when I get a response, but the company has already said that "our focus will be on scaling up Model S production over the coming year, so no major platform changes to the hardware are planned in the near term."

(Notice that the letter to shareholders doesn't explicitly say the car will change lanes automatically. Instead, it talks about 'automatic cruise control' and 'highway steering', whatever that may mean).

Why does this matter?

Automatic lane change is not the most 'advanced' feature announced by Tesla. Its own blog post says "imagine having your car check your calendar in the morning ... calculate travel time to your first appointment based on real time traffic data, automatically open the garage door with Homelink, carefully back out of a tight garage, and pull up to your door ready for your commute." Not content with this, Tesla states that it "could also warm or cool your car to your preferences and select your favority news stream." Phew!

However, these features are preceded by words like imagine and could, and they haven't been demoed yet. By contrast, automatic lane change was the most prominent item in the Autopilot event. It has been shown in several company-produced videos. And yet, Tesla has never mentioned that the driver may need to check the adjacent lane himself.

Without discussing the legal implications of this feature, let's think about how it would work. First, sonar checks the adjacent lane - but if you actually want to move two lanes or more, the driver will have to do it because sonar doesn't reach that far. Then the driver makes sure no car is coming, or at least not too quickly - again the sensors cannot do it. Then, he taps the turn signal. Then, the car steers itself - since the sensors can't reach two lanes away, we suppose that it will automatically change only to the immediately adjacent lane, and not further.

How does this make any sense?

UPDATE: several commenters have claimed that of course it's the driver's responsibility to check the adjacent lane. While Tesla has not gone out and said the opposite, videos from the Autopilot event always show the lane change feature without any mention of this supposed responsibility. In fact, in the videos the car drives itself without touching the wheel!

www.youtube.com/watch

www.youtube.com/watch

Automatic lane change may be impossible even with a rear radar

The big scoop comes from Bertel Schmitt at Daily Kanban. He talked to Ken Koibuchi, Toyota's (NYSE:TM) head of autonomous vehicle development, about the possibilities of radars and autonomous cars. I won't quote as I think you should read the entire piece, but I'll summarize.

First of all, you absolutely need a rear radar - sonar and cameras don't cut it. Second, no radar currently on the market can do the task.

Hold that thought. Toyota's top autonomous car guy thinks automatic lane change is impossible right now. Of course, it may be possible in one of those experimental cars we see at CES, but not in the ones you see at a dealership. Or maybe it's possible if "automatic" for you means "the driver checks the other lanes because the sensors can't do it."

Hence, if Tesla is going to enable automatic lane change before October 2015, the company must really aim for the ultra-high-end in the radar market.

There is no Autopilot Ultimate

Or Home, or Plus, or Professional, or Tier I. There is simply Autopilot. Tesla has never talked about different levels within Autopilot features, so we can only assume that when such a rear radar is available, it will be available on all Model Ss - whether they are $70,000 or $140,000.

It seems that such a super-advanced radar would be a very expensive piece of equipment, and with gross margins of about $25,000 per car it could eat up much of Tesla's earnings. Is the company going to charge for it separately? Is it going to be part of the Tech package? If it turns out the feature will not be included in the basic Autopilot package, wouldn't the promotion of automatic lane change constitute misleading advertising?

I have also inquired about these things (well, about the first two). If you don't see an update here, it's because there was no response.

Would Tesla announce a project and then forget about it?

In fact it's something of a tradition.

  • In October 2012, Tesla claimed the solar panels installed on Supercharger canopies would generate more electricity than the Model S consumed, a statement that baffled many an analyst and PV aficionado. More than two years later, only a handful of Superchargers are equipped with said panels. And the company never reveals how much power they generate, or who is buying its power and for what price.
  • In June 2013, Tesla demoed a 90-second, fully automated, live-streamed battery swap. A year and a half later exactly zero swap stations have been built and no customer has ever been able to do what was advertised in the video - although that hasn't kept the company from getting swap-generated ZEV credits. UPDATE: Tesla appears to be building a battery swap station at Harris Ranch, but the company is almost mum on the issue and I've seen no confirmed deadlines or even a press release acknowledging the existence of such a station.
  • In August that year, Elon Musk announced the Hyperloop in the Tesla blog - here's the official proposal hosted on Tesla's website. While not a company project per se, it's still telling that Elon Musk would tease this project for over a year to deliver - 16 months after the announcement - nothing.
  • Then in October, Tesla put up the CHAdeMO adapter on its website with a 'Coming Soon' message. Fourteen months later it's still Coming Soon. It was released in Japan back in September, but owners are reporting it doesn't work half the time.
  • In its 2013 Q4 earnings release, Tesla stated that "In 2013, we developed stationary energy storage products for use in homes, commercial sites and utilities. The applications for these battery systems include backup power, peak demand reduction, demand response and wholesale electric market services. We plan to ramp sales of these products in 2014" (page 8). Needless to say, as 2014 draws to a close not a single commercial customer of these 'stationary energy storage products' can be named, though the company continues to hype its energy storage business - in magazines, not in the 10-Qs.
  • In June this year, Tesla stated that all its patents were available for other companies to use - not through a legally binding document, but through a blog post. Six months later, no company has been reported to use Tesla's patents.

Of course, these projects may just be waiting for their turn. Perhaps Tesla will start to install massive photovoltaic systems at every Supercharger site, and the Hyperloop will start construction and so on.

Just let me remind you that, when it comes to Autopilot features, Tesla talked about several months.

Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.

About this article:

Expand
Author payment: $35 + $0.01/page view. Authors of PRO articles receive a minimum guaranteed payment of $150-500.
Tagged: , , , Auto Manufacturers - Major,
Want to share your opinion on this article? Add a comment.
Disagree with this article? .
To report a factual error in this article, click here