You won't be meeting the big, red Baxter robot anytime soon in your local supermarket. But the soft-surfaced, slow-moving industrial robot known for its safety around humans has been programmed to work a checkout lane and choose how it accomplishes certain tasks.
Rethink Robotics' Baxter can already work a simple pick-and-place production line, and it can do so even more accurately after its recent 2.0 software upgrade. A Cornell University team led by doctoral candidate Ashesh Jain, under the tutelage of assistant professor of computer science Ashutosh Saxena, programmed a modified Baxter robot to work as a checker.
Click the photo below to check out our slideshow.
With multiple joints, a Baxter robot can move more flexibly than a human. Since it would be hard for a human to decide how to use those arms, the robot is programmed to plan its own movements and then allow humans to make corrections. As a first step, it computes three possible trajectories for moving an object and displays them on a touchscreen. After an operator selects one, the robot goes through the motions of that trajectory, and the operator can make refinements. (Source: Cornell University)
Like a human checker, the robot had to handle objects with different sizes, shapes, and fragility. A carton of eggs can't be tossed around like a box of cereal. Also, a sharp object like a knife shouldn't be moved in a trajectory that takes it too close to a human.
The Cornell University Personal Robotics Lab has been working on "coactive learning," helping humans and robots cooperate to accomplish tasks. An essential part of this approach is a robot's ability to adapt to people; it must be able to make some decisions on its own. The team taught Baxter to plan its own actions and gave it feedback in multiple iterations.
This is possible partly because Baxter 2.0's movements can be programmed using Waypoints. These let users define the specific trajectories of Baxter's arm movements very precisely by modifying a given trajectory while it's in progress to avoid obstacles or to place items in a certain way. Users train Baxter by pushing its arms into each point along the path and then pushing a button to set those points in memory.
As shown in this YouTube video, the Cornell researchers give corrective feedback to teach Baxter to use shorter trajectories to keep knives away from people. Baxter then knows which of several trajectories should be preferred for a given object in a given context. The team gave similar feedback for handling a carton of eggs; the robot learned not to lift it too high off the counter and to keep it upright and horizontal.
AnandY, I agree about the robot's adaptability. It's designed to be an industrial robot that's safer for humans than the current ones, although it doesn't have anywhere near their speed and force. Although the supermarket app was designed primarily for R&D purposes, it's pretty convincing for an end-application.
The approach in Baxter 2.0' programming, wherein the robot can learn new ways to do things throughout its life (within the limits its joints allows) means that this robot has the potential to be adapted for almost any job that is purely mechanical and does not need any artificial intelligence.
This is pretty impressive. The fact that you can set specific trajectories depending on the object being moved and the environment in which its being moved means that Baxter wont just cost people jobs, it might cost many of the present day assembly line robots their jobs as well. They are simply not as flexible as Baxter.
Reducing cost and programming complexity while increasing capability is the real news here:
"What used to cost $50k and required programming by experts now costs $20k and can be setup by a relatively unskilled user in 1/x the time."
The problem is that un-informed journalists start to drink the cool-aid and run-away with the speculative uses that the promoters of the technology champion:
Journalist/blogger: "What can you do with this technology?"
Scientist/Engineer: "It will change the world as you know it..."
One of my favorite examples of over-hyping is the "Segway". It was going to revolutionize how people got around. Parking problems and intercity traffic would be greatly reduced. Urban planning would have to take this exciting new technology into consideration when designing new public spaces. Did the journalist's come up with all these ideas or were they provided them by the Genius's that devised the technology? How much excitement would there have been on this product if instead it was pitched as a great way for tourists to see more of the locales they are visiting or for Police to patrol urban neighborhoods?
It is the inventor's job to think of all the possible problems that can be solved by their technology. It is the Journalists responsibility to vet the "pie in the sky" scenarios and ask the hard questions. But pragmatism won't get you page views.
Yes, I think Baxter has many more practical applications than Amazon's drone-delivery service, at least at the moment. Baxter is already becoming a proven technology whereas Amazon has some technology and logistical hurdles to face.
You're quite welcome, William. I think bagging would be more difficult for Baxter to do than checking, although the 2.0 software upgrade we wrote about here http://www.designnews.com/author.asp?section_id=1392&doc_id=268173 lets him move objects vertically as well as horizontally. That upgrade's Pick and Place capability would allow Baxter to pick items off a shelf and place them in a box, movements that resemble bagging. But as Chuck points out, it's complicated because of the different weights and fragility of items involved.
Cadman-LT, interesting idea about programming a checkout robot to not handle some items, like a knife. I agree with you about not trusting them much myself. However, like a production line, the focus is on speed and accuracy combined. Just like in a production line, where you don't want special stations humans have to carry something to for inspection, having to wait for a human to arrive to bag a knife or other exception item would slow the line down. Like waiting for someone ahead of you with a zillion coupons or an expired credit card.
Last thing about this. It's one thing if the bot glitches and breaks your eggs. It is an entirely different matter if he accidently slices your face! Scary to me! They shouldn't be allowed to handle dangerous objects. I still love the idea though! Ohh untill I get smacked in the head with a sack of potatos!...lol
In this new Design News feature, "How it Works," we’re starting off by examining the inner workings of the electronic cigarette. While e-cigarettes seemed like a gimmick just two or three years ago, they’re catching fire -- so to speak. Sales topped $1 billion last year and are set to hit $10 billion by 2017. Cigarette companies are fighting back by buying up e-cigarette manufacturers.
Advertised as the "Most Powerful Tablet Under $100," the Kindle Fire HD 6 was too tempting for the team at iFixit to pass up. Join us to find out if inexpensive means cheap, irreparable, or just down right economical. It's teardown time!
The increased adoption of wireless technology for mission-critical applications has revved up the global market for dynamic electronic general purpose (GP) test equipment. As the link between cloud networks and devices -- smartphones, tablets, and notebooks -- results in more complex devices under test, the demand for radio frequency test equipment is starting to intensify.
Much of the research on lithium-ion batteries is focused on how to make the batteries charge more quickly and last longer than they currently do, work that would significantly improve the experience of mobile device users, as well EV and hybrid car drivers. Researchers in Singapore have come up with what seems like the best solution so far -- a battery that can recharge itself in mere minutes and has a potential lifespan of 20 years.
Some humanoid walking robots are also good at running, balancing, and coordinated movements in group settings. Several of our sports robots have won regional or worldwide acclaim in the RoboCup soccer World Cup, or FIRST Robotics competitions. Others include the world's first hockey-playing robot and a trash-talking Scrabble player.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.