As part of its mGripAI platform, Bedford, Mass.-based Soft Robotics Inc. designed a new gripper with a host of sensors that can better assess what it is holding.
“What we did was embed force sensors in the fingers,” said Harley Green, director of business development at Soft Robotics. “Now we can tell if we have a product in the gripper for grip confirmation, and we can also tell if we’ve delivered product.”
The new gripper can also determine how strong of a grip it has on items and can adjust its acceleration to accommodate, Green said.
“We can actually machine learn on the products,” he said. “If we are moving something like chicken flats, we know they are a little bit more rigid than a chicken breast, so we can actually machine learn on the product, so when the product is in a certain zone of the gripper, we’re going to have an optimal grasp. We can change our grasp quality profile based on the product and the optimal gripping location.”
Return to article.