MIT’s New Hardware Tool Aims To Democratize On The Job Training For Robots

There’s no silver bullet when it comes to teaching robots. Future automation will learn and improve from some combination methods, including reinforcement and imitation learning, simulation, large datasets, and on the job training.

The growing number of training paradigms aim to improve accuracy and speed up processes. They’re also working to lower the barrier of entry for robot training. To achieve ubiquity at work and in the home, these systems cannot be gate kept. Users need options beyond learning Python or dialing up the friendly neighborhood roboticist for help.

A team of MIT engineers are among those looking to democratize access to robot learning. The versatile demonstration interface (VDI) combines three separate teaching methods into a piece of hardware that attaches to a robot arm like an end effector. Users can train the system through tele-op, manually move the arm, or detach the tool to demonstrate a task.

ROI Calculator

Discover the potential cost savings of robotic automation over a 20-year system life

This calculator compares your current manual labor costs against the total cost of owning and operating a robotic system over its 20-year lifespan.

EXPLORE TODAY

The researchers note that access to different training models addresses different user preference, as well as a variety of scenarios. Handling toxic waste and other dangerous substances is floated as an example where remote training is preferred. A worker in the same factory may prefer a more hands on approach in training the system to box up a product.

“We are trying to create highly intelligent and skilled teammates that can effectively work with humans to get complex work done,” says MIT’s Mike Hagenow, “We believe flexible demonstration tools can help far beyond the manufacturing floor, in other domains where we hope to see increased robot adoption, such as home or caregiving settings.”

The VDI includes a camera for tracking and pressure sensors to determine the level of force being applied. The team hopes the device will increase access to robot training in both factories and homes.