stub New Tool Improves Robot Grippers for Manufacturing - Unite.AI
Connect with us


New Tool Improves Robot Grippers for Manufacturing

Updated on
Image: University of Washington

A team at the University of Washington has developed a new tool that can design a 3D-printable passive gripper and calculate the best path to pick up an object. The new development could help improve assembly-line robots. 

The system was tested on 22 different objects, including a doorstop-shaped wedge, a tennis ball, and a drill, and it proved to be successful for 20 of the objects. Two of the objects successfully picked up were the wedge and a pyramid shape with a curved keyhole, which are usually difficult for multiple types of grippers. 

The research is set to be presented on Aug. 11 at SIGGRAPH 2022. 

Adriana Schulz is senior author and a UW assistant professor in the Paul G. Allen School of Computer Science & Engineering. 

Creating Custom Tooling for Manufacturing Lines

“We still produce most of our items with assembly lines, which are really great but also very rigid. The pandemic showed us that we need to have a way to easily repurpose these production lines,” said Schulz. “Our idea is to create custom tooling for these manufacturing lines. That gives us a very simple robot that can do one task with a specific gripper. And then when I change the task, I just replace the gripper.”

Objects have traditionally been designed to match a specific gripper since passive grippers can’t adjust to fit the object they’re picking up.

Jeffrey Lipton is co-author and a UW assistant professor of mechanical engineering. 

“The most successful passive gripper in the world is the tongs on a forklift. But the trade-off is that forklift tongs only work well with specific shapes, such as pallets, which means anything you want to grip needs to be on a pallet,” said Lipton. “Here we're saying ‘OK, we don't want to predefine the geometry of the passive gripper.' Instead, we want to take the geometry of any object and design a gripper.”

There are many different possibilities for a gripper, and its shape is usually linked to the path the robot arm takes to pick up the object. When a gripper is designed incorrectly, it risks crashing into the object when attempting to pick it up, which the team set out to solve. 

Milin Kodnongbua is lead author and was a UW undergraduate student in the Allen School at the time of the research. 

“The points where the gripper makes contact with the object are essential for maintaining the object's stability in the grasp. We call this set of points the ‘grasp configuration,'” said Kodnongbual. “Also, the gripper must contact the object at those given points, and the gripper must be a single solid object connecting the contact points to the robot arm. We can search for an insert trajectory that satisfies these requirements.”

UW team creates custom grippers to help machines pick up more things

Designing New Gripper and Trajectory

To design a new gripper and trajectory, the team first provides the computer with a 3D model of the object and its orientation in space. 

“First our algorithm generates possible grasp configurations and ranks them based on stability and some other metrics,” Kodnongbua said. “Then it takes the best option and co-optimizes to find if an insert trajectory is possible. If it cannot find one, then it goes to the next grasp configuration on the list and tries to do the co-optimization again.”

The computer outputs two sets of instructions once it finds a good match. The first is for a 3D printer to create the gripper, and the second is with the trajectory for the robot arm following the printing and attachment of the gripper. 

The team tested the new method on various objects.

Ian Good is another co-author and a UW doctoral student in the mechanical engineering department. 

“We also designed objects that would be challenging for traditional grasping robots, such as objects with very shallow angles or objects with internal grasping — where you have to pick them up with the insertion of a key,” Good said. 

The team performed 10 test pickups with 22 shapes. For 16 shapes, all 10 of the pickups succeeded. Most shapes had at least one success, and two did not.

Even without any human intervention, the algorithm developed the same gripping strategies for similarly shaped objects. This has led the researchers to believe that they could be able to create passive grippers that pick up a class of objects rather than a specific object. 

Alex McFarland is an AI journalist and writer exploring the latest developments in artificial intelligence. He has collaborated with numerous AI startups and publications worldwide.