General-purpose robots are hard to train. The dream is to have a robot like the Jetson’s Rosie that canperforming a range of household tasks, like tidying up or folding laundry. But for that to happen, the robot needs to learn from a large amount of data that match real-world conditions—that data can be difficult to collect. Currently, most training data is collected from multiple static cameras that have to be carefully set up to gather useful information. But what if bots could learn fr...
