Forum for Science, Industry and Business

Sponsored by:     3M 
Search our Site:

 

Ask the crowd: Robots learn faster, better with online helpers

27.06.2014

Sometimes it takes a village to teach a robot.

University of Washington computer scientists have shown that crowdsourcing can be a quick and effective way to teach a robot how to complete tasks. Instead of learning from just one human, robots could one day query the larger online community, asking for instructions or input on the best way to set the table or water the garden.


The UW’s robot builds a turtle model.

U of Washington

The research team presented its results at the 2014 Institute of Electrical and Electronics Engineers International Conference on Robotics and Automation in Hong Kong in early June.

“We’re trying to create a method for a robot to seek help from the whole world when it’s puzzled by something,” said Rajesh Rao, an associate professor of computer science and engineering and director of the Center for Sensorimotor Neural Engineering at the UW. “This is a way to go beyond just one-on-one interaction between a human and a robot by also learning from other humans around the world.”

... more about:
»Automation »Conference »action »complete »humans

Learning by imitating a human is a proven approach to teach a robot to perform tasks, but it can take a lot of time. Imagine having to teach a robot how to load the dishwasher – it might take many repetitious lessons for the robot to learn how to hold different types of cookware and cutlery and how to most efficiently fill the machine.

But if the robot could learn a task’s basic steps, then ask the online community for additional input, it could collect more data on how to complete this task efficiently and correctly.

“Because our robots use machine-learning techniques, they require a lot of data to build accurate models of the task. The more data they have, the better model they can build. Our solution is to get that data from crowdsourcing,” said Maya Cakmak, a UW assistant professor of computer science and engineering.

The research team, led by professors Rao and Cakmak, also includes UW computer science and engineering graduate student Michael Jae-Yoon Chung and undergraduate Maxwell Forbes. The team designed a study that taps into the online crowdsourcing community to teach a robot a model-building task. To begin, study participants built a simple model – a car, tree, turtle and snake, among others – out of colored Lego blocks. Then, they asked the robot to build a similar object. But based on the few examples provided by the participants, the robot was unable to build complete models.

To gather more input about building the objects, the robots turned to the crowd. They hired people on Amazon Mechanical Turk, a crowdsourcing site, to build similar models of a car, tree, turtle, snake and others. From more than 100 crowd-generated models of each shape, the robot searched for the best models to build based on difficulty to construct, similarity to the original and the online community’s ratings of the models.

The robot then built the best models of each participant’s shape.

This type of learning is called “goal-based imitation,” and it leverages the growing ability of robots to infer what their human operators want, relying on the robot to come up with the best possible way of achieving the goal when considering factors such as time and difficulty. For example, a robot might “watch” a human building a turtle model, infer the important qualities to carry over, then build a model that resembles the original, but is perhaps simpler so it’s easier for the robot to construct.

“The end result is still a turtle, but it’s something that is manageable for the robot and similar enough to the original model, so it achieves the same goal,” Cakmak explained.

Study participants generally preferred crowdsourced versions that looked the most like their original designs. In general, the robot’s final models were simpler than the starting designs – and it was able to successfully build these models, which wasn’t always the case when starting with the study participants’ initial designs.

The team applied the same idea to learning manipulation actions on a two-armed robot. This time, users physically demonstrated new actions to the robot. Then, the robot imagined new scenarios in which it did not know how to perform those actions. Using abstract, interactive visualizations of the action, it asked the crowd to provide new ways of performing actions in those new scenarios. This work will be presented at the Conference on Human Computation and Crowdsourcing in November.

Other research teams at Brown University, Worcester Polytechnic Institute and Cornell University are working on similar ideas for developing robots that have the ability to learn new capabilities through crowdsourcing.

The UW team is now looking at using crowdsourcing and community-sourcing to teach robots more complex tasks such as finding and fetching items in a multi-floor building. The researchers envision a future in which our personal robots will engage increasingly with humans online, learning new skills and tasks to better assist us in everyday life.

This research was funded by the U.S. Office of Naval Research and the National Science Foundation.

###

For more information, contact Cakmak at mcakmak@cs.washington.edu or 206-685-5643 and Rao at rao@cs.washington.edu or 206-914-4719.

Michelle Ma | Eurek Alert!

Further reports about: Automation Conference action complete humans

More articles from Power and Electrical Engineering:

nachricht Another Milestone in Hybrid Artificial Photosynthesis
31.08.2015 | Lawrence Berkeley National Laboratory

nachricht New high energy density automotive battery system from Fraunhofer IISB and international partners
25.08.2015 | Fraunhofer-Gesellschaft

All articles from Power and Electrical Engineering >>>

The most recent press releases about innovation >>>

Die letzten 5 Focus-News des innovations-reports im Überblick:

Im Focus: An engineered surface unsticks sticky water droplets

The leaves of the lotus flower, and other natural surfaces that repel water and dirt, have been the model for many types of engineered liquid-repelling surfaces. As slippery as these surfaces are, however, tiny water droplets still stick to them. Now, Penn State researchers have developed nano/micro-textured, highly slippery surfaces able to outperform these naturally inspired coatings, particularly when the water is a vapor or tiny droplets.

Enhancing the mobility of liquid droplets on rough surfaces could improve condensation heat transfer for power-plant heat exchangers, create more efficient...

Im Focus: Increasingly severe disturbances weaken world's temperate forests

Longer, more severe, and hotter droughts and a myriad of other threats, including diseases and more extensive and severe wildfires, are threatening to transform some of the world's temperate forests, a new study published in Science has found. Without informed management, some forests could convert to shrublands or grasslands within the coming decades.

"While we have been trying to manage for resilience of 20th century conditions, we realize now that we must prepare for transformations and attempt to ease...

Im Focus: OU astrophysicist and collaborators find supermassive black holes in quasar nearest Earth

A University of Oklahoma astrophysicist and his Chinese collaborator have found two supermassive black holes in Markarian 231, the nearest quasar to Earth, using observations from NASA's Hubble Space Telescope.

The discovery of two supermassive black holes--one larger one and a second, smaller one--are evidence of a binary black hole and suggests that supermassive...

Im Focus: What would a tsunami in the Mediterranean look like?

A team of European researchers have developed a model to simulate the impact of tsunamis generated by earthquakes and applied it to the Eastern Mediterranean. The results show how tsunami waves could hit and inundate coastal areas in southern Italy and Greece. The study is published today (27 August) in Ocean Science, an open access journal of the European Geosciences Union (EGU).

Though not as frequent as in the Pacific and Indian oceans, tsunamis also occur in the Mediterranean, mainly due to earthquakes generated when the African...

Im Focus: Self-healing landscape: landslides after earthquake

In mountainous regions earthquakes often cause strong landslides, which can be exacerbated by heavy rain. However, after an initial increase, the frequency of these mass wasting events, often enormous and dangerous, declines, in fact independently of meteorological events and aftershocks.

These new findings are presented by a German-Franco-Japanese team of geoscientists in the current issue of the journal Geology, under the lead of the GFZ...

All Focus news of the innovation-report >>>

Anzeige

Anzeige

Event News

Networking conference in Heidelberg for outstanding mathematicians and computer scientists

20.08.2015 | Event News

Scientists meet in Münster for the world’s largest Chitin und Chitosan Conference

20.08.2015 | Event News

Large agribusiness management strategies

19.08.2015 | Event News

 
Latest News

Siemens sells 18 industrial gas turbines to Thailand

01.09.2015 | Press release

An engineered surface unsticks sticky water droplets

01.09.2015 | Materials Sciences

New material science research may advance tech tools

01.09.2015 | Materials Sciences

VideoLinks
B2B-VideoLinks
More VideoLinks >>>