This website uses cookies primarily for visitor analytics. Certain pages will ask you to fill in contact details to receive additional information. On these pages you have the option of having the site log your details for future visits. Indicating you want the site to remember your details will place a cookie on your device. To view our full cookie policy, please click here. You can also view it at any time by going to our Contact Us page.

Machine learning robots who cook perfect hot dogs

24 February 2020

An advance in machine learning allows two robots, named Jaco and Baxter, to make a hot dog.

Shutterstock image

Teaching robots to perform complex tasks involves a framework that could apply to a host of tasks, like identifying cancerous spots on mammograms or better understanding spoken commands to play music. But first, as a proof of concept, they’re making franks.

Researchers still don’t fully understand exactly how machine-learning algorithms work. That blind spot makes it difficult to apply the technique to complex, high-risk tasks such as autonomous driving, where safety is a concern. In a step forward published in Science Robotics, Calin Belta, professor in the Boston University College of Engineering, and researchers in his lab taught two robots to cook, assemble, and serve hot dogs together.

Their method combines techniques from machine learning and formal methods, an area of computer science that is typically used to guarantee safety, most notably in avionics (electronics for aircraft) or cybersecurity software. These disparate techniques are difficult to combine mathematically and to put together into a language that a robot will understand.

Belta, a professor of mechanical, systems, and electrical and computing engineering, and his team employed a branch of machine learning known as reinforcement learning. When a computer completes a task correctly, it receives a reward that guides its learning process. Although the steps of the task are outlined in a “prior knowledge” algorithm, how exactly to perform those steps isn’t. When the robot gets better at performing a step, its reward increases, creating a feedback mechanism that pushes the robot to learning the best way to, for example, put a hot dog in a bun.

Integrating prior knowledge with reinforcement learning and formal methods is what makes this technique novel. By combining these three techniques, the team can cut down the amount of possibilities the robots have to run through to learn how to cook, assemble, and serve a hot dog safely.

In the video, systems engineering graduate researcher Guang Yang and mechanical engineering graduate researcher Zachary Serlin teach robots Jaco and Baxter to work together to safely cook, assemble, and serve a hot dog to a human.

Source: Liz Sheeley for Boston University

The original article can be found on Futurity


Print this page | E-mail this page

Maxon Group UK