I agree its a pretty weak article but its a blog, not a paper, so I would not expect it to rise to the level of rigor for something intended for one of the IEEE journals.
That being said, if you've ever heard of 'charity burnout' there is something similar called 'robotics burnout.' You discover it, you engage in excitement, but nothing happens, you can only be excited (or charitable it seems) so long before you need to take a break. But lets start at the beginning shall we?
So in 1985 the Homebrew Computer Club was waning, it had been the center of attention between 75 and 85 but with people like IBM in the game and Apple going much more 'corporate' it was more of a users group than the folks who were changing the world.
One of the special interest groups from that meeting was the Robotics SIG. A guy named Dick Prather who was active in that SIG decided that even if the computer group was dying, robotics was just getting started and so he did the SIG equivalent of a setsid(2) call and made the Homebrew Robotics Club an independent organization. It has met continuously since then (yes 26 years).
One of the things about 'robots' that most people don't get, is that fundamentally a robot is any machine that has some level of program-ability that does one or more tasks while adapting to its environment. Your dishwasher is a good robot, it washes the dishes for you, or the pots and pans, or the stem ware, it uses a variety of sensors to decide if the dishes are clean yet and it dries them afterwards. If you didn't have it you would be getting a sore back moving dishes from the counter-top to the sink, to the drying rack.
Folks have argued that to be a robot it would have to do it like humans do, but that is an angels-on-pins sort of argument. Generally robots are the expression of automation, and they have (as a market) been growing where ever it makes economic sense. As processors get cheaper and more powerful more and more things make economic sense.
Of course for things that are really expensive or really dangerous its really easy to justify the cost. So for things like disarming bombs, or hunting people down in a country you are not technically at war with and killing them are both easily justified costs if you can automate them.
ISRobotics, the guys who make the Roomba, make most of their money selling robots for things like mine clearing and recon, and yes shooting people. Founded in 1990 they were making $1M/yr in 1996 with 16 employees[1], that wasn't particularly sustainable but in 2001, with 9/11 they demonstrated the value of their packbots. And now at $401M/yr they are doing quite well except that over 60% of their revenue is "G&I" which is code for "Government and Industrial".
Now that isn't all bad, its just that instead of comparing the robotics revolution to the 'PC' revolution you have to compare it to the 'computer' revolution, which is to say that a whole lot of investment and development is focused on corporate and government use until the the industry can supply enough automation to slake their thirst and to give some time to 'regular' folks.
So in the 80's we had companies like Androbot and Heathkit approaching mobile robotics as gimmicks/educational, in the 90's we had a lot of toys and the start of robot (really armored R/C vehicles with minimal automation) combat. And now in the second decade of the 21st century we're seeing prototype self driving cars, a number of walking designs, and energy systems that can sustain things for more than a couple of minutes.
Bottom line, things are going along, and getting better, and in many ways getting better faster now than they have in the past, but there is so much eco-system that has to develop around building mobile robotics that can do useful work that the rate of change feels much more evolutionary than revolutionary.
When we look at robotics startups today, places like Willow Garage and Anybot you can get a good feel for how cool things could be and how far we are from seeing the kind of uptake like we did with personal computers.
Generally robots are the expression of automation, and they have (as a market) been growing where ever it makes economic sense. As processors get cheaper and more powerful more and more things make economic sense.
Yeah that's kind of it in a nutshell. The PC revolution was easy to see because computer hardware has been performing the same fundamental task for decades: computation. Memory, storage, processing, and I/O; that's it. All computers are basically the same in that sense. But that one fundamental task is so useful that virtually everyone benefits.
Because of this, an entire industry has focused on improving those 4 basic components of computing, making them ever cheaper and more capable. And as the capability increases, the utility increases as well.
It's not the same with robots. Robots are more likely to evolve gradually over time. There won't ever be one single thing you can point to and claim "that was the robot revolution" and it's going to be a lot slower because there won't be a Moore's Law for robotics.
That being said, if you've ever heard of 'charity burnout' there is something similar called 'robotics burnout.' You discover it, you engage in excitement, but nothing happens, you can only be excited (or charitable it seems) so long before you need to take a break. But lets start at the beginning shall we?
So in 1985 the Homebrew Computer Club was waning, it had been the center of attention between 75 and 85 but with people like IBM in the game and Apple going much more 'corporate' it was more of a users group than the folks who were changing the world.
One of the special interest groups from that meeting was the Robotics SIG. A guy named Dick Prather who was active in that SIG decided that even if the computer group was dying, robotics was just getting started and so he did the SIG equivalent of a setsid(2) call and made the Homebrew Robotics Club an independent organization. It has met continuously since then (yes 26 years).
One of the things about 'robots' that most people don't get, is that fundamentally a robot is any machine that has some level of program-ability that does one or more tasks while adapting to its environment. Your dishwasher is a good robot, it washes the dishes for you, or the pots and pans, or the stem ware, it uses a variety of sensors to decide if the dishes are clean yet and it dries them afterwards. If you didn't have it you would be getting a sore back moving dishes from the counter-top to the sink, to the drying rack.
Folks have argued that to be a robot it would have to do it like humans do, but that is an angels-on-pins sort of argument. Generally robots are the expression of automation, and they have (as a market) been growing where ever it makes economic sense. As processors get cheaper and more powerful more and more things make economic sense.
Of course for things that are really expensive or really dangerous its really easy to justify the cost. So for things like disarming bombs, or hunting people down in a country you are not technically at war with and killing them are both easily justified costs if you can automate them.
ISRobotics, the guys who make the Roomba, make most of their money selling robots for things like mine clearing and recon, and yes shooting people. Founded in 1990 they were making $1M/yr in 1996 with 16 employees[1], that wasn't particularly sustainable but in 2001, with 9/11 they demonstrated the value of their packbots. And now at $401M/yr they are doing quite well except that over 60% of their revenue is "G&I" which is code for "Government and Industrial".
Now that isn't all bad, its just that instead of comparing the robotics revolution to the 'PC' revolution you have to compare it to the 'computer' revolution, which is to say that a whole lot of investment and development is focused on corporate and government use until the the industry can supply enough automation to slake their thirst and to give some time to 'regular' folks.
So in the 80's we had companies like Androbot and Heathkit approaching mobile robotics as gimmicks/educational, in the 90's we had a lot of toys and the start of robot (really armored R/C vehicles with minimal automation) combat. And now in the second decade of the 21st century we're seeing prototype self driving cars, a number of walking designs, and energy systems that can sustain things for more than a couple of minutes.
Bottom line, things are going along, and getting better, and in many ways getting better faster now than they have in the past, but there is so much eco-system that has to develop around building mobile robotics that can do useful work that the rate of change feels much more evolutionary than revolutionary.
When we look at robotics startups today, places like Willow Garage and Anybot you can get a good feel for how cool things could be and how far we are from seeing the kind of uptake like we did with personal computers.
[1] http://money.cnn.com/magazines/fortune/fortune_archive/1996/...
[2] http://investor.irobot.com/phoenix.zhtml?c=193096&p=irol...