Hacker News new | comments | show | ask | jobs | submit login
Autonomous transportation launching to 125k residents in Florida (voyage.auto)
157 points by olivercameron 39 days ago | hide | past | web | favorite | 67 comments

Hello everyone! Oliver, the CEO of Voyage and YC alum (2011) here.

I got started in the self-driving car world as a member of the team that built Udacity's self-driving car curriculum, and also an open source self-driving car. This industry is crazy fun and moving ridiculously fast.

Let me know if I can answer any questions at all about absolutely anything.

We're really excited about bringing Level-4 transportation to the world, but especially so at places like The Villages, Florida. Whether it’s helping those with Parkinson’s, Alzheimer’s, vision impairment, or just those who want to get around with less friction, we’ve seen first-hand the positive impact autonomous transportation can have for seniors.

Hi Oliver, a few questions:

(1) Will these cars have a Voyage employee in them to supervise?

(2) What are your driver intervention rates, eg. compared to Waymo? [1]

(3) Is there anything unique about The Villages, FL that makes it easier (or harder) to deploy self driving cars?

Congrats and interested to see the responses!

[1] https://www.usatoday.com/story/money/cars/2017/01/13/waymos-...

Hi marcell, great questions.

1) Yes, at least today. We call them Operations Specialists, and they're highly trained in operating autonomous vehicles and customer service.

2) This is a tough one, since it's not apples-to-apples. That said, Waymo is clearly the industry leader in this regard, and we're glad that at least some of their formula is observable from the outside. We've been at this a little less than a year, and today many of our passenger rides are 95+% autonomous. A long way to go, though!

3) We love these communities for a number of reasons:

A) Speed limits in these communities tend to be a little more restricted than everyday public road. The faster you (and others) can go, the more complex it becomes. The roads are also spectacularly well-maintained.

B) Customer acquisition is a non-trivial problem, and with these sorts of partnerships we gain (exclusive) access to 125,000 potential riders. It would be brutal for us to achieve the same in San Francisco, and not necessarily cost effective to retain them.

C) If we wanted to try out concepts that may accelerate the timeframe to remove the Test Driver, a private community is easier to collaborate with.

D) Above all else, there's just a lot of demand for what we offer in these sorts of communities. Self-driving cars will lower the cost of transportation, and for those with vision impairments (or Parkinson’s, Alzheimer’s etc.), it enables such residents to get around with less friction.

> 2) today many of our passenger rides are 95+% autonomous

Do you mind if I ask what portion of your rides are ~100% autonomous? I feel like that's a more interesting metric for progress toward removing human operators.

Hi Oliver. Good luck with this great project.

>> (1) Will these cars have a Voyage employee in them to supervise?

>>>> 1) Yes, at least today. We call them Operations Specialists, and they're highly trained in operating autonomous vehicles and customer service.

My understanding of the autonomy levels (for what they're worth) is that while a human is behind the wheel, autonomy is not at level 4. If a human is required to take control in certain situations, it's either level 3 or 2.

I have a feeling human operators are there for more customer service purposes. Given the location selected it might be wise to have staff on hand. The idea of no humans makes me think of the news story recently on the front page here about the hitchhiking robot being destroyed also seems relevant, although private areas may be less prone to incident like this. (https://www.wired.co.uk/article/hitchbot-usa-vandalised-phil...)

Also according to this techrepublic article, while level 3 may still require a driver, a level 4 vehicle can still be limited to the "operational design domain". So human drivers can still be in place for extreme edge cases. Although I don't know the specifics of these vehicles, and don't think the "levels" of driving are very useful.

Edit, forgot link: https://www.techrepublic.com/article/autonomous-driving-leve...

Thanks for the info. One more question if you have the time: to what degree are you limited by data, vs. other factors? If you could have 10x the training data, would that 10x your performance, or are other factors in the way (eg. compute, neural net architecture, etc.)?

We design our algorithms to not be necessarily reliant on more data for better performance. We recognize that fleet size and large-scale data collection just isn't going to be a key strength for Voyage, so we design our algorithms accordingly. For example, our classification stack required only 9,000 point cloud frames to achieve pretty amazing performance.

The Villages is basically a retirement community so there's a large audience of potential customers that may not be able to drive on their own.

Also, the max speed limit on roads in the community is only 30mph and electric or gas golf carts are the ubiquitous transportation method today.

Hi Oliver, congratulations on this! I have a few questions:

1) Is there a plan to be further integrated with the car eg hiding the LIDAR from plain sight?

2) On OSS, can you direct me to any OSS projects that is related to control the car ie the part from computer to the car or connecting the CANBUS.

3) What are the technical considerations, specific to Autonomous car, since you’re using OSS projects for it?

1) It's not an immediate priority for us, and not really a strength. We are partnering with an amazing company for our second generation vehicle that will go a long way towards this, though. In general, we are excited to see LIDAR sensors become slimmer.

2) One of our engineers builds and maintains Carloop.io, which I can verify as awesome!

3) Safety and performance. One project to keep an eye on is ROS2, which is prioritizing development of real-time features (RTOS, DDS support).

> We are partnering with an amazing company for our second generation vehicle that will go a long way towards this, though. In general, we are excited to see LIDAR sensors become slimmer.

I bet I can guess who this is


I can't, help?

I grew up near The Villages. Really crazy to see such a cool innovation basically coming to my home town. That being said, I don't know if self-driving cars are the right approach in that market. Have you considered self-driving golf carts? =P

(for those of you who don't know, The Villages has arguably more miles of golf cart paths than highways, and all shopping plazas have golf-cart sized parking spots, so most of the retirees cruise around on their carts)

Hi Oliver, this is a really cool announcement. I have some questions:

1) What are some of the issues surrounding adoption in local municipalities that you've faced with the deployment of this, specifically in Florida?

2) How do you go about convincing a local municipality to change their laws or issue waivers regarding legal issues about vehicles on the roadway that are autonomous, as specifically in Florida, the legal framework does not exist yet that I'm aware.

3) Explain your pricing model for the taxi's in The Villages and also does pricing decrease once the vehicles become fully autonomous (no driver in the front seat required)

I'd personally like to see a day, in my lifetime, where the pricing model of autonomous transport rivals that of vehicle ownership (which in my math is about $0.20 - $0.50/mile)

re: "Whether it’s helping those with Parkinson’s, Alzheimer’s, vision impairment, or just those who want to get around with less friction..."

I can see the positives of less friction for this segment of the population. Kudos.

That said, what effort has Voyage put into helping the local community (and law enforcement) prepare for this?

For example, a person with Alzheimer's gets lost...now what? True, this might happen now but if you're successful then the social infrastructure might not be ready, willing and able to handle the add demands on that system.

I'm going to hang up and take my answer off the air :)

We take our responsibility with any passenger _very_ seriously. We strive to be world-class in how we operate 100% of the time, and we learn everyday from the best customers we could find as to how we should handle situations like these.

No one has figured this out yet, so we have some ways to go, but rest assured we take our responsibility super seriously.

But this isn't simply a passenger issue. There's implications to the broader social fabric.

I agree, no one has figured it out yet. But I'm not so sure anyone has tried. Action <> Reaction is not an "algorithm" tech companies excel at.

I think a push button that will bring the car to the original place where the passenger ride would be nice.

Congratulations on your project Oliver.

How far off do you think it is from a regulatory perspective before you can have the cars drive without anybody sitting in them?

We’re ready! You can do those sorts of rides today (totally driverless) entirely legally in Florida and Arizona.

This was what Google was thinking for their little 25MPH bubble car. The one they discontinued in 2016.[1] The great thing about a 25MPH speed limit is that slamming on the brakes deals with most problems. You don't have a stopping distance long enough that you have to steer out of them.

Good idea, still too expensive per unit. Votage has a huge upfront cost spread over very few cars. Plus human drivers, which means "it doesn't really work yet". This is a "can we get market share before the money runs out" play. It worked for Uber. Things like this will be in retirement communities everywhere around 2025, once the sensors are in volume manufacture and cheap.

When I was working on self-driving (2005 Grand Challenge) I saw two initial use cases - this one, and airport rental car pick-up and return. Both require only slow-speed driving in reasonably controlled environments, something which can be made to work. Neither is yet economically feasible.

[1] https://www.technologyreview.com/s/603126/google-puts-the-br...

> The great thing about a 25MPH speed limit is that slamming on the brakes deals with most problems.

Might still not be that great if your clients are elderly.

On suburban roads, nearly all obstacles will be seen early enough to allow for slow breaking. It's very different in cities where people can appear between parked cars.

I meant as passengers. Restraint systems are designed with healthy bodies in mind, and they wreak havoc on the elderly. "Slamming the breaks" is not a valid strategy in this environment.

I think this business model has a far greater chance of succeeding than something like Tesla's. It's easier to just focus on a small averagely populated area than high density urban cities or in Tesla's case, the entire country. It's safer to deal with construction/road issues, unique quirks present of each city when you're only focusing on one community at a time.

It's far less of a headache dealing with one set of regulations for a community that has few, impressionable decisionmakers than trying to scale while trying to meet ever-changing individual regulations from each state/city and an uncertain national landscape. Only Google and the big automakers have the available influence to push for federal regulation standards and even those can be scrapped if some major incident happens or morons refuse to follow proper safety protocols to chase profits (uber).

If Tesla pushes level 4 or 5 autonomy, they need to consider every circumstance (rain, snow, dust storms, tornadoes, ice, and so many other random unique Mother Nature quirks), every road construction issue, hazards, fallen trees etc. all at once. It just isn't feasible. You need city employees to report fallen trees, garbage truck routes, area with frequent deer or animal crossings etc. They can't just say it's autonomous in one particular area where maybe only 100 Teslas are present, it's just not economical.

I think this route of small community adaptation is the most likely and it's great to see most players shift to this model. This means that automakers aren't going to be the big players in this space for another decade or so. It's not economical for big beuracratic, slow moving businesses like Ford to enter into this landscape. Small companies and startups pushing into this has its advantages and disadvantages however.

I'd just like to point out that Tesla as you're speaking of it is a car manufacturer, whereas Voyage is a car service.

Tesla is apparently planning to launch their own car service, which from an economic standpoint may be poised to win. It will involve Tesla owners being able to add their car to a public fleet and essentially have their car make money for them.

It's plausible the Tesla model will win out because, in the majority of hours, only about 20% of the fleet is used in a car service. RideAustin, as a nonprofit, has publicly available ride data to support this. So with a dedicated fleet, you'll often have a lot of unused cars that will be quite inefficient. With the Tesla model, however, the money-making portion is a secondary purpose for the car's ownership, with the primary one being the owner's own convenience when needed.

Hope this is explained accurately!

If you can afford a Tesla why would you let your car be used like this?

I am the kind of person who would get a tesla and I would never want to deal with the inconvenience of lending my car to a public fleet so it could get fucked up by drunk people.

Also how does the car know if someone throws up in it before picking up another passenger?

The car arrives, the passenger opens the door, turns away in disgust, loads the ride hailing app and presses the "report vomit" button. The car drives itself to the cleaning service, another car turns up shortly.

In the end, how long did you wait for a car?

I'm a strong believer in self driving cars but don't think car ownership will decline meaningfully. Having your own self-driving car still brings you home when you're drunk but without waiting in the cold (or surge pricing if the bar just closed).

Don’t know why you were downvoted. I fully agree. Last thing I want is my car interior abused like public transportation. Litter, scrap food, lice, body fluids. No thanks.

> Also how does the car know if someone throws up in it before picking up another passenger?

Camera + microphone + vomit recognition software.

What if it made you an extra $6000 per month? Even $3000?

Obviously the market would adjust from current taxi rates, but this is based off of the following rough estimate:

20 hours per day of you not driving * 30 days * 11$ avg taxi take-home per hour[1] = $6600

[1]: https://money.usnews.com/careers/best-jobs/taxi-driver-and-c...

Most car owners today could save money by owning a different car, reducing the cars in the household or using more public transport.

Owning a car is (for most) not just a necessity, it's an expression of freedom. Being reliant on a car sharing app is not ideal for everyone, esp. families.

How much does the car cost?

Also I don't think it would last very long driving around 20 hours/day every day.

It would be like airbnbing my apartment. Huge pain in the ass all the time with maintenance and dealing with other related issues.

I think the calculus is more that people won't own their own cars, because it will cost 10k/year to own, but 2k/year to buy trips.

> It's plausible the Tesla model will win out because, in the majority of hours, only about 20% of the fleet is used in a car service. RideAustin, as a nonprofit, has publicly available ride data to support this. So with a dedicated fleet, you'll often have a lot of unused cars that will be quite inefficient. With the Tesla model, however, the money-making portion is a secondary purpose for the car's ownership, with the primary one being the owner's own convenience when needed.

Problem: Peak demand for car service customers and peak usage for owners is likely to be at the same or overlapping times.

Sure, there's always going to be that. But for people that work different shifts than the normal 9-5, it could be a boon for them. It wouldn't have to be a majority to have a huge impact on transportation and people's livelihood.

Thank you! We think so too.


Especially if there are serious competitors in this space, governments could enact laws that speed up safety requirements (and maybe speed up innovation too).

One thought: Governments could force driverless companies to share their data, in some inter-operable format (think a 3D map, or 6D if you had speed information?) that could be read by other cars nearby. For instance if there is a stroller with a baby in front of the car in front of you, or if an accident happened in the street you are about to enter, you would be able to read it from the information emitted from cars that are closer to the danger/obstacle. Current technologies allow this. My guess is that sensors will gain substantially in terms of robustness. (Imagine what you can do with one set of lidar/thermocamera/visial cameras; now imagine what you can do with data from 10 of these sets from different locations!)

Smaller players (such as Voyage, although it seems you got there far already!) would probably benefit from this and it would benefit competition/innovation. Governments would likely be happy to force this into law because if it makes sense for safety.

Anyway, congrats for the real-life launch

I haven't heard of Voyage before, but this is interesting.

Would it be accurate to summarize that while most autonomous car companies are going for gradually higher levels of autonomy nationwide/worldwide, Voyage has the opposite strategy of starting at a higher level of autonomy in one controlled community and will aim to expand geographically from there?

Correct! We think people would be surprised at how big some of these private communities are, and how many places look like them other than retirement communities.

In robotic terms it's called a constrained environment.

With a human behind the wheel, with hands off? Or are they really launching with nobody in the vehicle?

Human behind the wheel today. A key reason we choose private communities is that we see it as a way to remove that Test Driver much faster, though.

This picture from the article has a human behind the wheel: https://cdn-images-1.medium.com/max/2000/1*Hz0b8qPI6Q4pwysyt...

With a test driver behind the wheel. Source: https://twitter.com/olivercameron/status/951200429001469953

the picture looked like no one in the vehicle, but then again there was this in the past: https://i.kinja-img.com/gawker-media/image/upload/s--o8x0LC0... :-)

"Beginning in early 2018, we’ll start rolling out a door-to-door self-driving taxi service to residents."

Now? Isn't that a bit optimistic? To me it looks like they have just one car for testing. At least they understand that they need the environment mapped to a very fine degree.

We are an optimistic bunch, and it will be a slow rollout (constrained by vehicle build-outs), but we also have experience launching such self-driving taxi services.

We have a service that's live today in San Jose that serves tons of (repeat) riders daily. We've learned a lot there that we're applying directly to Florida.

How autonomous is the service in San Jose then? Do you mean to say that you have cars without drivers that serve ride after ride and come home to charge? If so I'm far behind the curve.

We still have Test Drivers behind the wheel as a last resort, but we regularly give passenger rides with zero or low disengagements.

As a real-world example: we have a couple who uses our self-driving taxi service multiple times a day every day (The Grims), and their rides are ~95+% autonomous.

You mean 95% of their rides are entirely autonomous or their typical ride is 95% autonomous?

I remember how voice to text got to 95% in the nineties. It's used for a lot of things these days, still not for safety-critical tasks.

A typical ride is 95% autonomous.

A long way to go for sure, but having been at this just less than a year, we're feeling good!

So how do you get rid of the human backup driver? Until they go, this technology has negative cost-effectiveness. Remote manual control at slow speed for difficult situations? Dispatching a backup driver?

Relying on fine-grained maps always seems to me like the Wrong Way to do these things. Fine-grained maps are guaranteed to be wrong, so you can't rely on them. And if you don't rely on them then you don't need them.

It depends on how fine-grained we're talking. If you rely on your map for absolute perception (every object must exist in your map in its current real-world state), then yes, you're in trouble. If you rely on the map as a basic prior, and perceive the world through different means, then a map (to me) is the right way to go. You avoid a whole category of computer vision in doing so.

The good news is that a point cloud map is relatively trivial to create, and the pipelines to update such a map over time are becoming more understood. There's lots of startups taking on this challenge.

It's how you start. At some point computers will have general cognition at a level that enables them to understand typical traffic situations and improvise. Right now the road network is not standardized enough for them to rely on their sensors. One way streets? Need a plan. Turn regulations? Plan! Interpretation of lights? Hah, plan. Go over a curb? Not in the plan!

They simply can't take those decisions. Yet. As companies such as this make headway, roads will become more standardized to help them along.

Pretty cool idea to go to private, well constrained areas so that it scopes down the problem and allows for quick iteration. Good luck! Excited to see autonomous vehicles in these environments.

Thank you Dhruv!

This sounds really a great project. Good luck.

I wonder there was any consideration given to just adding automnomy as a module to the golf-carts that are currently widely used in these communities. It seems that could allow for faster deployment, lower legal barriers and lower capital investment.

The autonomous vehicles in Black Mirror's recent Hang the DJ fits this concept.

I had no idea that retirement communities could grow to the size of cities. This place is amazing [1] [2]. Peter Thiel and Y Combinator talk about building cities or planned communities, but it looks like it's already been done. And now they're adding fully autonomous vehicles. I think this is also the perfect place to test fully automated grocery stores, restaurants, and home automation. You could set up automatic payments for everything via facial recognition / GPS / bluetooth. I never considered that the "city of the future" would be a retirement community with lots of elderly people.

I wonder if they have security cameras covering every street, similar to the UK. You could set up some facial tracking across all the cameras, and a centralized security system in every house, and it might become the safest city in the world. It could also make a great episode of Black Mirror.

I work remotely on my laptop, and my wife and I have a lot of hobbies, including yoga, ballet, archery, art, music [3]. We're living in Chiang Mai, and many of the archery club members are retirees from the UK. So we're already interacting out with some older people, and don't have any qualms about it. All residents of The Villages must be over the age of 19, but 20% of the houses can be owned by younger people who live by themselves. Not saying we would want to live there, but it's not too hard to imagine.

This would also be a great place to build home automation startups (or even city automation), just like Voyage is doing. For example, many of the residents might be able to afford robotic chefs [4]. The system could automatically order your groceries, and they'd be delivered by a self-driving Voyage car. Then another robot in your house could receive the delivery and store the food in your smart refrigerator.

It also makes me realize that a blockchain might be unnecessary. You trust your neighbors, you trust the people who manage the community, and you trust the legal system, lawyers, judges, and prisons. So there's probably no need for a trustless payment network in a community like this. The Octopus card works fine in Hong Kong, and it's just integers in a database.

Lots of people are working from home these days, so I think this might lead to more planned communities. The digital nomad lifestyle is great for a while, but it gets exhausting after a few years. Hopefully there will some communities that are geared towards younger people and people with families. I've also been following the High Street Cohousing Project [5] in New Zealand. Would be very interested to hear about other cohousing projects and planned communities.

[1] https://www.thevillages.com

[2] https://en.wikipedia.org/wiki/The_Villages,_Florida

[3] https://www.thevillages.com/calendar/

[4] http://www.moley.com/

[5] http://highstreetcohousing.nz/

I grew up 2 miles from The Villages and watched it blossom into the huge, beautiful community it is today. It was something amazing to see. I really think it's one of those untold business success stories that folks outside of that area are not too familiar with.

Although it was a little awkward growing up in a town with so many senior citizens, it was really nice in some ways. The Villages brought amenities to our small town that we would have never had access to.

However, I would argue The Villages is a lovely place to live and a fantastic place to retire, but it's not very suitable for tech startup. There is no technical talent there. Florida as a whole suffers from a lack of a great tech ecosystem (although there's a few big cos. here and there). Of the working age population, The Villages mostly consists of service and health workers.

That being said, you might be able to find/attract some talent in Orlando or Gainesville. Still, it's nothing like what's in some of the major tech hubs around the U.S.

Perhaps a less obvious benefit is financial. No schools saves a lot of property tax.

The Wikipedia post seems to indicate that they are still zoned to school districts.

Applications are open for YC Summer 2018

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact