VC Funding for Robotics in 2015 | Hacker News Reader

VC Funding for Robotics in 2015

hizook.com

46 pointsbeambot17 comments

Who's profitable? Once upon a time, 1983 was the year of the robotic startup. That didn't end well.[1] It looked so promising back then; mapping using ultrasonic sensor was starting to work. But it wasn't good enough.

So far, the only profitable autonomous products are industrial robots and vacuum cleaners. (The surgical robots like DaVinci are teleoperators.) Even Boston Dynamics, in the end, failed to produce a usable product; the LS3 (the militarized version of BigDog) was canceled.

The technology is much better today than ever before, and there's a lot that could be done. But so far, there's nothing that's cost effective enough to be produced in volume. Except robot vacuum cleaners.

If robot vacuum cleaners were cost effective and able to keep out out of trouble (such as getting snagged in cords) most offices would be using them. That's not happening yet.

[1] https://www.frc.ri.cmu.edu/~hpm/project.archive/robot.papers...

Of these? I don't know.

Across robotics... Lots! Many of them are in manufacturing (eg Kuka, Schunk, etc). There are lots in the military (eg UAVs in the big defense companies). There are lots in automation (eg Festo and Parker Hannifin). There are in logistics (eg. Kiva, before being acquired by Amazon). There are lots in healthcare (eg intuitive surgical). And its pretty clear that autonomous cars are on the horizon at this point.

A lot has changed since 1983... We're seeing a resurgence due to the confluence of major improvements in several key components: processors, displays, batteries, connectivity, sensors, and algorithms (including AI and ML). Unsurprisingly, many of these are being driven by Moore's Law and Cellphones. You can think of a robot as a "Cellphone with actuators" and you wouldn't be too far off. The actuators haven't changed too much since 1983... but the cellphone part has.

Human labor is a $60 trillion dollar market. Robots that are capable of general learning will be able to capture a good portion of that.

It's ok to lose money developing such technology. There's a long road ahead, and we should be impressed with those willing to invest now.

Are drones considered robots? If so then they're a smashing success.

Some level autonomy is a requirement to call something a robot. Lots of drones are remote controlled and thus not (fully) autonomous and not robots in my eyes.

Not on the list and also not profitable yet is all the money pouring into development of self driving cars. They are robots too, aren't they?

Maybe the funding from Google, Apple, Uber, Volvo, GM etc. isn't venture capital, strictly speaking, but I would assume that it easily dwarfs most of the funding mentioned on the list.

Great to see this list on HN. It is a huge resource to the robotics community.

Great list. What do you think is driving the growth?

It's probably a combination of things... The medical robotics companies have always raised pretty healthily. The drone companies are seeing a lot of traction, as are some of the consumer robot companies.

But probably the biggest contributing factor: Major companies like Google, Toyota, and Amazon are spending $Billions each year on robotics efforts... and there are surprisingly few robotics startups for them to draw from. VCs are starting to realize that there are probably some real winners out there, and are actually embracing hardware again.

There have been massive improvements in AI in the last few years. Particularly things like machine vision, speech recognition, and reinforcement learning. AI is one of the major things that has been holding back robotics. It may be that it's time has finally come, or at least is very close.

At least in the flying-robot space, I think cheap batteries and sensors have been at least as important.

The sensors -- primarily accelerometers -- are etched in silicon, and benefit from a lot of the same cost and quality improvements as other microelectronics.

Batteries don't get better as fast as processors, but they do get better. About 7% per year for the past two decades. That has been adding up, and finally tipped us over the point where you can have your own tiny flying helicopter for $15.

It may requires great protection on intelligent property since there are too many copycats in the market. If they can "copy" with good quality, it would be more acceptable (although it is not right). But if they are only produce electronic rubbish as usual, they would destroy the industry and lower consumers' confidence with their low cost rubbish.

How has Sphero/Orbitix raised so much money? They have raised $88M total.

When will be the year when a robot will clean my room? I've been waiting 28 years. Please just make it happen, someone.

Are you dirtier than a Roomba can handle?

The thing is that a Roomba (or Neato, I have one and they work great) only vacuums the floor. I have a kitchen, toilet, furniture that needs cleaning and dusting as well every once in a while. I still have to do my own laundry, except for the actual washing itself.

I want a robot that does all my cleaning, not just the floor. Preferably without buying 10 different robots but one that does it all.

How effective is the Roomba? Most reviews that I've read say that it doesn't replace the need to vacuum yourself.

Reply on news.ycombinator.com