Scientists have showcased an ‘ambidextrous’ robot that can learn different ways of moving their arms grippers to pick up objects in the most effective way.
Researchers from the University of California, Berkeley, have created a system that allows robots with multiple arms to analyse and rapidly decide which grip would pick up an object most successfully.
The latest invention could help speed up processing in warehouses for e-commerce companies.
By determining which type of grip is the most likely to succeed, the robot is able to operate quickly on a feedback loop.
The system uses a ‘reward function’ assigned to each type of gripper which the robot then calculates the probability of each type of grip succeeding on a specific item.
The researchers said their two-armed robot had achieved a 95% reliability rate.
Jeff Mahler, the research’s lead author and a postdoctoral researcher at the university said: ‘Any single gripper cannot handle all objects.’
‘For example, a suction cup cannot create a seal on porous objects such as clothing and parallel-jaw grippers may not be able to reach both sides of some tools and toys.’
As e-commerce companies become increasingly reliant on machinery to cope with growing order demands, robotics could play important roles in future warehouses.