Issue link: https://iconnect007.uberflip.com/i/1228683
APRIL 2020 I SMT007 MAGAZINE 37 strations of setting the table with the objects. Then, the researchers tasked the arm with automatically setting a table in a specific configuration, in real-world experiments and in simulation, based on what it had seen. To succeed, the robot had to weigh many possible placement orderings, even when items were purposely removed, stacked, or hidden. The researchers' robot made no mistakes over several real-world experiments, and only a handful of mistakes over tens of thousands of sim- ulated test runs. (Source: MIT News Office) quotation purposes. Even before they get the job, they simulate it and say, "I want to charge you X plus 10% because your design is going to cost me with some challenges." Johnson: Sagi, thank you very much for the information. This has been very helpful. The big takeaway for me is in changing manage- ment's thinking much more than changing the equipment or the processes. You need to look at your business to make Industry 4.0 happen; it doesn't happen just by buying sensors and collecting data. Reuven: Exactly. I'm happy that I managed to express my mindset. Matties: Thank you very much. Reuven: Thank you. SMT007 simulate it after you fixed it a little bit. You can publish it to the next step, which is in the man- ufacturing area. Matties: Is this where the phrase "digital twin" comes into play? Reuven: Exactly. It is the digital twin. The idea from Siemens is that now you're able to have it on both sides—from the mechanical side and the electronics side. Matties: If you build it on the digital side first, then it becomes predictive, and you are pre- dicting the output. Reuven: Yes. Again, it can be very simple things. Can I test the board in an effective way? We found out that some of our users even use the DFM/process engineering tool; they use it for Roboticists are developing automated robots that can learn new tasks solely by observing humans. In the work- place, you could train robots like new employees, showing them how to perform many duties. Making progress on that vision, MIT researchers have designed a system that lets these types of robots learn complicated tasks that would otherwise stymie them with too many confusing rules. One such task is setting a din- ner table under certain conditions. At its core, the researchers' "Planning with Uncertain Specifications" (PUnS) system gives robots the humanlike planning ability to simultaneously weigh many ambiguous—and potentially contradictory— requirements to reach an end goal. In doing so, the system always chooses the most likely action to take, based on a "belief" about some probable specifications for the task it is sup- posed to perform. The researchers 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 demon- Showing Robots How to Do Your Chores