Udacity spins out self-driving taxi startup Voyage

Image jpg

UPDATE: Sebastian Thrun answers student questions for 24 minutes in this video Self-Driving Car Nanodegree: Q&A with Sebastian Thrun. This video is probably the most informative insider perspective on the fast-moving autonomous vehicle space.

One example of how fast the field of AI is moving: Udacity’s “school for robo-cars has been so successful that it’s now spinning out of Udacity into its own company, Voyage.” Here’s a snippet from Business Insider:

(…snip…) The new spin-out will be lead by Oliver Cameron, a Udacity VP that was spearheading a lot of its self-driving car curriculum. The company broke the news to its employees Wednesday morning.

Udacity will have a stake in the newly-formed company as part of the deal, said the Udacity’s CMO Shernaz Daver. Voyage also recently closed a seed round of funding that included Khosla Ventures, Initialized Capital, and Charles River Ventures.

Voyage has been hot in Silicon Valley investor circles because of one big name linked to Udacity: Sebastian Thrun. Thrun, who founded the education startup, is also nicknamed the “Godfather of self-driving cars” for the work he did at Google and helped launch the self-driving car nanodegree program at Udacity.

Thrun, though, says he’ll have no connection with Voyage even though it’s spinning out of his company. “Because of personal conflicts, I have excused myself from any involvement in Voyage. I wish Oliver and his team all the best,” Thrun said in a statement to Business Insider.

The autonomous taxi startup wants to bring about the end goal where autonomous cars can carry people anywhere for a very low cost, Cameron said. It already has permission to deploy its self-driving cars to ferry passengers in a few places over the next few months, but Cameron declined to specify where.

“We want to deploy these not within five years, but very soon. We think in terms of weeks, not in terms of years or months,” he told Business Insider in an interview.

Pure guess: one reason Oliver Cameron decided to take this risk is because Udacity is open-sourcing it’s own self driving car project. All the code is there for all of us to use and improve. Including the new startup Voyage. And Oliver Cameron has a pretty good idea how successful the Udacity project is going to be.

Update: this week BMW has announced they plan to ship self driving cars in four years, in 2021. That’s similar to plans already announced by GM, Ford, Chrysler, Mercedes, Volvo and Chinese ride-sharing giant Didi Chuxing.

Elon Musk forecasts Tesla will self-drive from NY to LA in ~2 years

Elon Musk sent this pair of tweets this morning. The Summon/Unsummon commands are supposed to work today on your iPhone or Tesla key.

Elon Musk (@elonmusk) 1/11/16, 09:08
Tap your phone or key and your car will open the garage door, exit, close the door and come to you. Will do same in reverse for unsummon.

Elon Musk (@elonmusk) 1/11/16, 09:11
In ~2 years, summon should work anywhere connected by land & not blocked by borders, eg you’re in LA and the car is in NY

Now we know why Tesla is developing the robot arm that plugs their fast charger into your Tesla charge port 🙂 So you can park your car in NY and tell it to pick you up in LA. It will just need to navigate to the string of cross country Fast Charging Stations. Story and robot arm video at Wired.

Tesla robot charger

DWD Driving While Distracted: Google skips auto-pilot, goes for fully self-driving vehicles

At any given daylight moment across America, approximately 660,000 drivers are using cell phones or manipulating electronic devices while driving, a number that has held steady since 2010. — Distraction.gov Key Facts and Statistics

“I think it’s wonderful that Tesla has gone out there with this technology, but they might have hyped Autopilot a little bit too much. It doesn’t work in all circumstances. Drivers don’t necessarily know when the car goes from tracking fine to a gray area when the car is confused, and then to a situation when the car doesn’t know where it’s going. These things aren’t well-defined.” —  Alain Kornhauser, director of the transportation program at Princeton University

I’ve been puzzling over the question of how Tesla-type auto-pilot systems are really going to work in the real world. That’s the world where drivers are likely to turn over way too much responsibility to the auto-pilot. Drivers are already frequently distracted while they are theoretically in control. Googles test data makes it very clear that drivers are going to over-trust the auto-pilot brain, and therefore fail to take control fast enough, with enough context awareness to become the pilot. I’ve found that the monthly reports from the Google Self-Driving Car Project are a super research source: “what is working, what isn’t working, what probably won’t ever work.” Read their latest October 2015 report:  “Why we’re aiming for fully self-driving vehicles”. 

As we see more cars with semi-autonomous features on the roads, we’re often asked why we’re aiming for fully autonomous vehicles. To be honest, we didn’t always have this as our plan.

In the fall of 2012, our software had gotten good enough that we wanted to have people who weren’t on our team test it, so we could learn how they felt about it and if it’d be useful for them. We found volunteers, all Google employees, to use our Lexus vehicles on the freeway portion of their commute. They’d have to drive the Lexus to the freeway and merge on their own, and then they could settle into a single lane and turn on the self-driving feature. We told them this was early stage technology and that they should pay attention 100% of the time — they needed to be ready to take over driving at any moment. They signed forms promising to do this, and they knew they’d be on camera.

We were surprised by what happened over the ensuing weeks. On the upside, everyone told us that our technology made their commute less stressful and tiring. One woman told us she suddenly had the energy to exercise and cook dinner for her family, because she wasn’t exhausted from fighting traffic. One guy originally scoffed at us because he loved driving his sports car — but he also enjoyed handing the commute tedium to the car.

But we saw some worrying things too. People didn’t pay attention like they should have. We saw some silly behavior, including someone who turned around and searched the back seat for his laptop to charge his phone — while travelling 65mph down the freeway! We saw human nature at work: people trust technology very quickly once they see it works. As a result, it’s difficult for them to dip in and out of the task of driving when they are encouraged to switch off and relax.

We did spend some time thinking about ways we could build features to address what is often referred to as “The Handoff Problem”– keeping drivers engaged enough that they can take control of driving as needed. The industry knows this is a big challenge, and they’re spending lots of time and effort trying to solve this. One study by the Virginia Tech Transportation Institute found that drivers required somewhere between five and eight seconds to safely regain control of a semi-autonomous system. In a NHTSA study published in August 2015, some participants took up to 17 seconds to respond to alerts and take control of the the vehicle — in that time they’d have covered more than a quarter of a mile at highway speeds. There’s also the challenge of context — once you take back control, do you have enough understanding of what’s going on around the vehicle to make the right decision?

In the end, our tests led us to our decision to develop vehicles that could drive themselves from point A to B, with no human intervention. (We were also persuaded by the opportunity to help everyoneget around, not just people who can drive.) Everyone thinks getting a car to drive itself is hard. It is. But we suspect it’s probably just as hard to get people to pay attention when they’re bored or tired and the technology is saying “don’t worry, I’ve got this…for now.”

Regulators: please let us have our robocars fast. They won’t be flawless, but they will be much safer than the meatware currently driving steel lethal weapons around our cities. 

3 Weak Arguments Against Self-Driving Cars Totally Miss the Point.  Let’s not let ourselves get distracted from the main goal here. In the hour that you debate the “moral dilemma” at a dinner party almost 300 people were killed somewhere by driver error [WHO Road Traffic Deaths].

The electric car is going to take over the world. Soon.

I certainly hope that forecast proves correct. That post is from Y Combinator partner Geoff Ralston. In his post Geoff raises an interesting “tipping point” argument:
When 10% of the vehicles on the road are electric many of them will go out of business.
What is the logic that connects 10% EV penetration to “many gas stations out of business”? Other possibilities are the gas-pumping retailers get more physically concentrated (which has been happening in most geographies that we know); the retail price of petrol goes up; automated gas pumps become accessories to other retailers (supermarkets are common gas pumpers in NZ).
I agree with Geoff that consumers will react strongly when it becomes really inconvenient to refuel. It’s above my pay grade whether that is 25% or 50% EV penetration. When does the refueling “tipping point” effect grow large enough to offset the (decreasing) EV price premium?
What else might bend the curve of EV adoption? It wouldn’t surprise me if self-driving cars were a more direct cause of an inflection in EV market penetration. I think automated taxis are likely to explode in the cities that have favorable density, travel patterns and demographics. And I will eat my hat if the robo-taxi companies chose to deploy fleets of ICE vehicles. If I’m right about the growth of robo-taxis these fleets could contribute EV growth promoters like:
1. Building out fleets of Robo-taxis drive down the EV cost/benefit so much it promotes private adoption.
2. In urban zones the market for private autos shrinks as people find they prefer using over owning. So total EV penetration goes up with the shrinking ICE denominator.
Even if the EV takeover is slower than Geoff hopes we can look forward to a much cleaner and more convenient future. EVs are essential: the road to a low carbon economy goes through the electrification of transport.

Uber shares their data feeds with host cities

Click to embiggen

This is a very good move for Uber: it costs the company very little to share their data feed with each city where they operate. That data is hugely valuable to the city management. And it is a perfect way to incentivize cities to welcome rather than fight Uber’s presence.

(…snip…) As we have grown, so has our ability to share information that can serve a greater good. By sharing data with municipal partners we can help cities become more liveable, resilient, and innovative.

Today, Boston joins Uber in a first-of-its-kind partnership to help expand the city’s capability to solve problems by leveraging data provided by Uber. The data will provide new insights to help manage urban growth, relieve traffic congestion, expand public transportation, and reduce greenhouse gas emissions.

(…snip…) This data can be utilized to help cities achieve their transportation and planning goals without compromising personal privacy. By helping cities understand the way their residents move, we can work together to make our communities stronger. Smart Cities can benefit from smart data and we will champion municipal efforts devoted to achieving data-driven urban growth, mobility and safety for communities.

Morgan Stanley predicts driverless car ‘utopia’ by 2026

Morgan Stanley Research

Adam Jonas isn’t Nostradamus, but the Morgan Stanley analyst is predicting the road to complete vehicle autonomy will begin in 2026. What’s more, he says the technology will eventually reach 100 percent market penetration two decades later.

Wall Street forecasts aren’t generally superior to the dart-throwing monkeys, but they do create great charts. At end-2018 we can reality-test the Adam Jonas report: there should be at least a couple of full-autonomous offerings on the market.

The majority of media reports I see on robocars are focused on the idea that Danny-the-driver sells his Prius, buys his first robocar — but continues to drive to work every day. The “big change” is that now Danny can TXT while not-driving instead his usual TXTing-while-driving, otherwise known as DWD, Driving While Distracted. That will save a lot of lives, including those of cyclists like myself.

But that isn’t the significant revolution. The real revolution will be most obvious in high density cities like New York or San Francisco. High robocar penetration will happen there first because of the appeal of “whistle-cars”. We are already seeing this in the explosive growth of Uber, Lyft and similar services. Urbanites are proving they prefer to summon a just-in-time ride on their iPhone. Robocars will make this service even more convenient and a LOT cheaper. That’s pretty much the end of the self-owned urban car market.

Even earlier than the whistle-car we will see delivery-bot vehicles. This is where Amazon is going with their drone development program. The drones will be the “last block” of the delivery web. The delivery-bots will handle the larger, heavier delivery loads and supply the drones and the human powered deliveries.

American cities typically have 40 to 50% of their useable area eaten by automobiles, comprised of streets and parking. Most of that space can be released to productive use once robocars and delivery-bots reach full penetration. Perhaps cities will even allow building again – so the acute shortage of affordable housing can be eliminated. Or maybe not – the same old status-quo people will probably still control the city governments. And they already have their multi-million dollar positions – no “housing crisis” for them.

For a deep dive into the implications of robocars, I recommend Brad Templeton, now a consultant to Google’s robocar program. See Brad’s main robocar page: Where Robot Cars (Robocars) Can Really Take Us. There’s lots more here on Seekerblog

Germany’s gamble on P2G: Expensive electricity to expensive methane to expensive storage


There isn’t much good news about Germany’s failing Energiewende. Because they have boxed themselves into a corner by eliminating their nuclear portfolio, they are facing the physical realities of trying to deploy unreliable wind and solar far beyond their appropriate penetration levels. According to the research by Lion Hirth (Vattenfall Europe AG) appropriate penetration would be 7% wind for today’s Germany, perhaps as much 25% in a future Germany optimized for VRE [PDF]. Germany is abusing neighboring grids by dumping excess generation while exploiting the neighboring grids to supply power when the sun doesn’t shine and the wind doesn’t blow.

To push VRE to meet their 80% commitment Germany obviously needs abundant cheap storage. Unfortunately they don’t have the enviable volume of hydro that Sweden and Norway have. So, as  Quirin Schiermeier writes in Nature Renewable power: Germany’s energy gamble, Germany is making a big bet on P2G:

P2G, however, could provide a vast amount of new storage capacity and Germany is leading the way. The plant in Stuttgart has 250 kilowatts of electrolysis stacks, which use electricity from renewables to produce hydrogen from water. To make methane, the hydrogen is reacted with CO2 from decomposing sewage and agricultural waste at a nearby biogas plant. Other P2G plants could scrub CO2 from the air.

But P2G is still an immature technology, with high upfront costs and an efficiency of only about 50% in converting electricity to methane. Synthetic methane plants have also struggled with the purity of their product. At the ZSW facility, the main goal is to routinely produce gas with low oxygen and hydrogen content.

For transport fuels, the P2G concept might make some economic sense IF powered by nuclear electricity. But Germany is trying to use this process to fix the serious grid instability problems they have created for themselves. Adding the high cost of variable renewable electricity to the high cost of the P2G conversion – this is good for the economy? How prudent is it to base an industrial economy on advanced proton-exchange membranes for electrolysis?

Toyota Bets Against Tesla With New Hydrogen Car


Personal hero and hard-nosed engineer Elon Musk calls them “fool cells”. Contrast the Musk view to what seems to be a very big bet by very-smart-company Toyota. Here’s Fool.com “A big bet that puts Toyota at odds with Tesla“: Every time I examine the hype for the “hydrogen economy” I shake my head. Why would Toyota cancel their deal with Tesla while announcing a “new hydrogen fuel cell car (that) will start arriving at dealers sometime next year”?

The Fool quotes Bloomberg “the car is expected to be called the “Mirai,” the Japanese word for “the future”.” That’s appropriate because the “F” in FCV has so far stood for “10 years in the Future”.

In “Tesla Trumps Toyota: Why Hydrogen Cars Can’t Compete With Pure Electric Cars” Joe Romm argues the practical and economic superiority of BEVs over FCVs (Battery Electric Vehicles over Fuel Cell Vehicles). Good arguments — let me know what you think.

In the comments to Joe’s Energy Collective piece there are a number of useful comments. Especially this one by regular contributor Roger Arnold on hydrogen economics and footprint: 

(1) If you’re talking about the most economical and widely implemented production method for hydrogen (i.e., from reforming of natural gas), then the carbon footprint for the FCV is substantially worse than if the gas were used directly in an IC engine. You’ve gone to a lot of trouble and expense for a worse result.

(2) If, instead, you’re talking about the more expensive route of producing hydrogen by electrolysis of water using zero-carbon electricity, then you could get two to three times better mileage per kWh by using that electricity to charge batteries rather than make hydrogen.

The main potential advantages that FCVs can deliver over BEVs are driving range and fast refueling. But those are non-issues for the commuting and shopping trips that comprise the overwhelming bulk of miles driven. Going on a road trip? Then rent a gasoline vehicle for that purpose. With the coming era of autonomous vehicles, the rental agency will deliver the vehicle to your driveway, and drive it back to their lot after you’ve returned home.

And this comment by Nathan Wilson:

Fuel cell vehicles have momentum because of the hype. If we launched a petroleum phase out tomorrow, and hydrogen FCVs, BEVs, and ammonia ICE cars had to compete in the market, I would expect 80% of the sales to go to ammonia, 19% to BEVs, and 1% for hydrogen FCVs. This is mostly based on sticker price, but also on the much easier infrastructure situation for ammonia versus hydrogen (ammonia cars can be dual fuel with gasoline backup, but HFVC cannot; ammonia can be transported by truck, but H2 cannot).

The idea that any technology we like can be made cost competitive is appealing, but has no basis in reality.

Tyler Cowen: Tesla Says “All Our Patent Are Belong To You”

Tyler Cowen has some very big news from Tesla.

Elon Musk writes:

Yesterday, there was a wall of Tesla patents in the lobby of our Palo Alto headquarters. That is no longer the case. They have been removed, in the spirit of the open source movement, for the advancement of electric vehicle technology.

Tesla Motors was created to accelerate the advent of sustainable transport. If we clear a path to the creation of compelling electric vehicles, but then lay intellectual property landmines behind us to inhibit others, we are acting in a manner contrary to that goal. Tesla will not initiate patent lawsuits against anyone who, in good faith, wants to use our technology.

When I started out with my first company, Zip2, I thought patents were a good thing and worked hard to obtain them. And maybe they were good long ago, but too often these days they serve merely to stifle progress, entrench the positions of giant corporations and enrich those in the legal profession, rather than the actual inventors. After Zip2, when I realized that receiving a patent really just meant that you bought a lottery ticket to a lawsuit, I avoided them whenever possible.

…We believe that Tesla, other companies making electric cars, and the world would all benefit from a common, rapidly-evolving technology platform.

Technology leadership is not defined by patents, which history has repeatedly shown to be small protection indeed against a determined competitor, but rather by the ability of a company to attract and motivate the world’s most talented engineers. We believe that applying the open source philosophy to our patents will strengthen rather than diminish Tesla’s position in this regard.

I believe that this announcement will be discussed in business schools for years to come much like Henry Ford’s announcement of the $5 a day wage.


Uber is a Dating Service

It is very interesting how people with different frameworks react to events. An example is the media storm related to SF startup Uber.com and dynamic pricing. For the second year now Uber has utilized dynamic pricing for peak demand periods such as Halloween and New Years Eve. Perhaps the media will continue to make January the “dump on Uber” month; or maybe they will go back to easy fillers like “Ten best of 2013”.  Uber seems to be like the political-horse-race meme, journalists can generate column inches without even getting out of their chair. They just need a Twitter account. 

Back to frameworks. I will offer only two examples to illustrate:

  1. computer science, economics
  2. journalists

#1 When I read about Uber inside my framework #1, I see Uber as a dating service that leverages machine learning. They earn a fee only when they match-mate successfully, and better than competitors. They earn more fees if their algorithms allow them to better position their drivers where the demand will be (micro demand prediction) They don’t “own” either side of these matches. In particular, it’s the drivers that are really setting the price and agreeing to the pricing scheme. Uber is providing the platform to enable the transaction. So given my framework, what I see is:

  • Charges of “greed” and “gouging” don’t make sense – if the Uber service doesn’t attract drivers to Halloween or New Years Eve then there is no match. No ride for me, no fee for Uber.
  • If Uber is going to be the winner in this match-making competition, then their secret sauce is going to have to make both buyer and seller prefer Uber to the other ride matching services. Profiles of the “daters” are absolutely essential. Yet journos complain that “the driver rated the passenger”. 

#2 On hearing somebody paid 7x the typical fare on to get home in a December snowstorm? Obviously that is the Uber corporation taking advantage of the helpless passenger. The alternative of the stranded passenger in the snow doesn’t come to mind. On hearing a story that Uber didn’t tell the passenger what the fare would be until they get a big Visa bill they think “Of course, that’s what that greedy corporation would do”. The alternative that the passenger didn’t pay attention to all the Uber price notices and warnings isn’t considered.

I was first aware of the Uber dynamic pricing trials in 2012 from Joshua Gans’ post Uber and the delicate business of creating a platform. Reacting to the first media storm, Joshua summarized Uber’s challenge

Basically, to satisfy one side of its market — the taxi drivers — Uber upset the other side — its customers.

At the time I thought Uber would experiment until they found the right balance to satisfy both “daters”. So far it appears that this techno-optimist called it wrong. I’ve investigated as best I can the bill-of-accusations against Uber. None of it holds up. I’ve examined their FAQ, their user guide, and a number of Travis Kalanick’s blog posts on dynamic pricing. I’m impressed — objectively Uber has worked very hard to avoid surprising a customer with a high fare. How can a client complain “Uber didn’t tell me what the ride would cost” when, from NZ (!), I was able to obtain a London Uber quote in about 60 seconds on my iPad on a slow internet connection. Starting with finding where on the Uber site to get my quote, then choosing London as my city, then typing in my route from Mayfair to Canary Wharf. 


Further down the quote screen is the exact fare basis, which is the same time and/or distance fare basis as for most taxis, including Uber London competitor HAILO. In October Hailo’s minimum fare has gone up to £10 between 6am to 10pm Monday to Sunday; and £15 between 10pm to 6am.

And the famous London Black Taxis. I said same fare basis, not the same coefficients. 



If you read through the Uber Happy New Year bulletin published Dec 30, the day before NYE, you’ll see how totally clear the Uber UI on what you will pay — you can’t just press I ACCEPT HIGHER FARE under the Huge Type multiplier, you also have to key in the multiplier (7 in my example above). And the depth of the coaching (“pro tips” to avoid paying high fares). Travis begins with this:

New Year’s Eve is upon us and we want to give you some quick pro tips for getting around with Uber! This New Year’s Eve we’ll have a record number of cars on the road ready to get you where you want to go. But, that doesn’t change one simple fact: on NYE, everyone wants to move around the city at exactly the same time!

You can avoid the peaks of surge pricing with good timing when you travel. Check out our smart tips below, and don’t forget you’ll always know the price before you request.

So here’s my question: what can Uber do differently than the above, or this (from a Travis Kalanick Dec 16 email reply to “outraged client”):

We regularly do surge pricing when demand outstrips supply. Remember, we do not own cars nor do we employ drivers. Higher prices are required in order to get cars on the road and keep them on the road during the busiest times. This maximizes the number of trips and minimizes the number of people stranded. The drivers have other options as well. In short, without Surge Pricing, there would be no car available at all.

Now granted, that the prices are significantly higher. BUT we notify every customer in big bold images in text, which each customer has to confirm in order to request. Furthermore, every customer also had to type in what the multiplier was in order to double confirm that they understood what they were agreeing to.

So, was it expensive. It was, and we wish it wasn’t necessary. But if you did indeed take the rides described then you confirmed the price which was very up front, and then entered the multiple you read into a text box in order to double confirm.

Airlines and Hotels are more expensive during busy times. Uber is as well. We don’t just charge to make a buck though, we take a small fee of the transaction, but the vast majority goes to the driver so that we can maximize the number of drivers on the road. The point is in order to provide you with a reliable ride, prices need to go up.

If you have other ideas for how to provide a reliable ride during busy times, I am all ears. In the end, Uber is reliable, always, and we will create a system that maximizes the number of people that can get safe and reliable rides. Not surging is saying you shouldn’t have the option. Not surging is saying we should be just like a taxi and be unreliable when people need us most. These are outcomes that take choices away from the consumer and make it harder to get around cities – these are outcomes that we put a lot of hard work in to avoid so that at least you have the choice if you want one.