, 29 tweets, 6 min read Read on Twitter
Thread:
Uber has announced an effort to deploy autonomous scooters and bicycles.

This is the worst new #mobility idea yet.

If I can't convince you that this is an insane waste of time by the end of this thread, I shouldn't have my job.

techcrunch.com/2019/01/20/ube…
I accept that it is technically possible to design, engineer, and manufacture autonomous scooters and bikes.

But it's a terrible idea for so many reasons I don't know where to begin.
I'm just going to start listing as many of the many reasons I can think of that this the worst idea ever.

As I do, keep in mind I'm not dismissing the technical possibility. I'm dismissing the idea that there is a business model at the end of this rainbow.
Let's start with keeping these things upright. Difficult, but not impossible.

But will require some kind of complex precision-manufactured gyroscopic stabilization system.

That will require weight, cost, and complexity. A *lot* of complexity.
The development of the stabilization system will be co$tly. It may not even be feasible. Best case scenario, you're adding thousands of dollars of hardware to these things just to keep them upright.

That doesn't even get into the autonomy platform...
They will need sensors.

GPS, but that's easy.

More critically they'll need machine vision sensors giving at least a 180 degree view of the world, and more likely a 360 degree view. So we're talking a lidar and a couple cameras, maybe several cameras, who knows!
Now where the eff are you going to mount these things on a bike or scooter!? They need to be unobstructed and intrusive when there's a rider on. They also need a good view of the environment so looking handlebars or front post.
On bikes/scooters these parts turn in the direction of motion, which may or may not be useful, but adds complexity and it will be hard to adapt any of the existing vehicle-based vision algorithms for this purpose so they'll all have to be programmed de novo.
The chassis of bikes and scooters are also more subject to shocks, shakes, and changes in pose than a 4-wheel vehicle. This makes it much harder for the sensors to get a coherent image of the environment. Gyroscopic balancing will complicate this even further.
And speaking of those shocks and shakes, the sensors on this thing will have to be sturdier than automotive grade. I can't imagine bolting precision robotics systems onto a kick scooter and keeping them clean and in good repair for any amount of time.
If you somehow get your gyro-stabilizers and sensors onto this thing, you need to connect those to at least one ultra complex control unit. This will need every bit as much processing power as a self-driving car, maybe more.
Perhaps you can shrink your processing unit down to a smart-phone size. Perhaps you can fit it in a post or under a panel and insulate it from weather and vibrations.

You're still not done.
Self-driving cars get to utilize existing vehicle drive-by-wire technology.

There is no steer-by-wire technology for bikes and scooters because ... why would there be?

That is a whole 'nother engineering program just for that.

Plus more weight, more cost, more complexity.
So, assuming that Uber can get this to work, it will no longer be a utilitarian device. It will be some frankenscooter with gyroscopic stabilizers, multiple vision sensors, remarkably high-end compute unit of some sort, and steering actuators that have yet to be invented.
I won't even try to guess the value in hardware bolted on to these things to get them to work even poorly.

But I'll bet it's more than an order of magnitude beyond what makes sense for a dockless bike/scooter service.
Furthermore, even if this stuff works in simulation, and works in controlled physical trials, you really think this is going to work in meat space?

The real world? Have you *seen* the real world?

Let's imagine the day in the life of an autonomous scooter ...
Imagine:

You're a dockless autonomous bike or scooter (it barely matters which).

You were left on the sidewalk after a ride. You're battery is low and you're programmed to go and recharge...
I guess I'll assume that Uber has already worked out a way to transition from kick-stand supported to mobile (another non-trivial engineering project).

If you've fallen over, I guess you're done for. Unless they invent some kind of spider arms.
If you're attached to a bike rack, I guess you're done for. Really, there are many many situations in which the smartest robot scooter could not extricate itself from however an average person might leave it on the sidewalk.
But lets assume by luck or design you're in a position that allows you to automatically begin a journey to go get recharged ... WHERE?
Uber is going to buy real estate for charging stations? How many? The battery was already low (that's WHY it needs to recharge). So, do you have these things on every block? What happens when you get there? Are there robot arms that plug it in? Inductive charging? Who knows!?
And can you imagine these things teetering down a sidewalk? First of all, anyone who's spent any time as a pedestrian or bike knows that the infrastructure for us is far less maintained and controlled as the roads for vehicles.
Urban biking or scooting requires frequent lifts over curbs, across gutters, down stairs, through grass, etc. etc. etc.

The operation domain is far more difficult than that for a self-driving taxi, which Uber has not proven adept at creating.
Plus, c'mon... There's no way these things are going to pass more than a couple pedestrians before somebody kicks it over. (Again, a need for spider arms.)

If I saw one of these things teetering my way on a sidewalk I can't claim to be above fucking with it.
I've literally reached the thread limit but there are still more reasons this is a bullshit idea.

Hey, why don't people respond with your own reasons this is a bullshit idea!
There's a company in Singapore trying to do this with electric ... standing tricycles (?). I wasn't thinking about this as included in "scooters." This would make the engineering easier, but operations harder; less mobile, less maneuverable, heavier, less fun. And still $$$.
Uber may as well skip the foreplay and just present the ultimate form-factor for autonomous micromobility.

(Not that this is a bad thing, but stop with the self-scooting scooters bullshit.)
In fact, you could also have the front wheel be the drive wheel, and balance by turning it 180. It'd be hard to stop and go, but not impossible.
Missing some Tweet in this thread?
You can try to force a refresh.

Like this thread? Get email updates or save it to PDF!

Subscribe to 🚗E🚌P🚚D🚲
Profile picture

Get real-time email alerts when new unrolls (>4 tweets) are available from this author!

This content may be removed anytime!

Twitter may remove this content at anytime, convert it as a PDF, save and print for later use!

Try unrolling a thread yourself!

how to unroll video

1) Follow Thread Reader App on Twitter so you can easily mention us!

2) Go to a Twitter thread (series of Tweets by the same owner) and mention us with a keyword "unroll" @threadreaderapp unroll

You can practice here first or read more on our help page!

Follow Us on Twitter!

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!