MIT Robot Adjusts Grip on Objects Based on Surroundings

Robotic hands can be rather clumsy. With little more than pincers and claws, they are not exactly known for their dexterity.

With five fingers it’s quite easy for humans to adjust our grip on an object. However, this isn’t typically the case with industrial robots.

In response, MIT engineers have designed a robot algorithm that can adjust a machine grip using the environment to its advantage.

If the robotic arm picks up a rod at the wrong spot it doesn’t have to drop it and pick it up a second time. Instead, the robot can use the algorithm to slightly loosen its grip and push the rod against a nearby wall to slide it into the correct position. The researchers have dubbed this “extrinsic dexterity.”

Professor Alberto Rodriguez and graduate student Nikhil Chavan-Dafle have created an algorithm which can predict the optimal force to adjust the grip of a robot as it pushes the clasped object against its surroundings. The model tracks the interactions between the gripper, the clasped object and external fixtures. The model was then tested with a real gripping robot.

“The agreement was pretty good,” Rodriguez says. “We’ve validated the model. Now we’re working on the planning side, to see how to plan motions to generate certain trajectories. One of the things we want to ask in the future is, ‘How do you engineer fixtures in the environment so that a robot’s motions are more reliable and can be executed faster?’”

The new robotic algorithm has the ability to improve various industries including manufacturing, medicine, disaster response and more. Extrinsic dexterity can allow robots to do more complex movements without additional capital costs opening the door to shrink the size of assembly lines, due to the need of fewer robots and workers.

“Chasing the human hand is still a very valid direction [in robotics],” said Rodriguez. “But if you cannot afford having a $100,000 hand that is very complex to use, this [method] brings some dexterity to very simple grippers.”

The researchers are looking into other movements that can adjust their grip including tossing and catching an object or rolling the object over a surface. However, their current research focuses on “prehensile pushing,” using fixtures to move the grasped object.

“We’re sort of outsourcing that dexterity that you don’t have in the gripper to the environment and the arm,” Rodriguez explains. “Instead of dexterity that’s intrinsic to the hand, it’s extrinsic, in the environment.”

In a recent video, James Anderton, Director of Content at ENGINEERING.com said “People are still more flexible than machines. In modern high flexibility operations humans actually have a significant advantage over robots.” This flexibility that Anderton eludes to is all about how robots are limited to their programming, while human workers can adapt to changing scenarios and job requirements.

This constraint often limits a robot to replacing humans for single tasks. Anderton said, “One for one task replacement will rarely be cost effective in today’s low interest rate environment, given the high capital cost of general purpose robotics.”

With the inclusion of this dexterity model, perhaps it will be possible for the robots to perform multiple tasks in the near future. This brings robots closer to completely replacing human workers.

If this model can be programmed into the robots already on the line then the capital costs are minimal to non-existent. Theoretically, one robot can be repurposed to do multiple tasks as the robot reorients the objects it needs, instead of the assembly line moving the objects to a robot in a different position.

Rodriguez agreed having said, “Exploiting the environment is and will be important for robots and the research community … Any applications where you have limitations in terms of payload, cost, complexity or areas like manufacturing, surgery, field operations or even space exploration — whenever you have a gripper that is not dexterous like a human hand, this [method] gives you some of that dexterity.”