Autonomous robots will soon take over mundane tasks, like setting tables, but that requires training. MIT researchers have designed a system that enables robots to learn complicated tasks using a new “Planning with Uncertain Specifications” (PUnS) system. This gives them humanlike planning ability to simultaneously weigh many ambiguous requirements to reach an end goal. Read more for a video and additional information.
For this specific task, the research team compiled a dataset with information about how eight objects — a mug, glass, spoon, fork, knife, dinner plate, small plate, and bowl — could be placed on a table in various configurations. A robotic arm first observed randomly selected human demonstrations of setting the table with the objects and then the researchers tasked it with automatically setting a table in a specific configuration, in real-world experiments and in simulation, based on what it had seen.
- Who doesn't love robots? Introduce kids to the creative world of coding with the best educational STEM toys to foster their curiosities. Building,...
- Includes 847 LEGO pieces that kids can build and rebuild into 5 cool multifunctional models. The best and most popular robotics toys for 7-12 year old...
- Construct and code Vernie the Robot to dance, rock out on the Guitar4000, foster Frankie the Cat, interact with the Autobuilder, or explore a new...
The vision is to put programming in the hands of domain experts, who can program robots through intuitive ways, rather than describing orders to an engineer to add to their code. That way, robots won’t have to perform preprogrammed tasks anymore. Factory workers can teach a robot to do multiple complex assembly tasks. Domestic robots can learn how to stack cabinets, load the dishwasher, or set the table from people at home,” said first author Ankit Shah, a graduate student in the Department of Aeronautics and Astronautics (AeroAstro) and the Interactive Robotics Group.
