Ken Goldberg, a robotics researcher at UC Berkeley and co-founder of Ambi Robotics, explains at TEDxMarin why household robots struggle with manipulation and describes progress in warehouse sorting and home tasks.
Goldberg contrasts impressive robot demonstrations with the frequent failures researchers see in practice. Moravec’s paradox helps explain the gap: tasks that seem simple to people, such as stacking blocks, can be difficult for robots, while some tasks that challenge people are easier for machines.
Complex robot hands can be heavy, costly, and unreliable, so Goldberg favors basic two-finger grippers and suction cups. Robot manipulation also suffers from uncertainty in motion control, sensing, and physical contact: cameras and lidar can miss important details, while tiny surface differences can change how an object moves.
E-commerce warehouses need to pick and sort many different items, a job still largely done by workers. Goldberg’s team developed Dex-Net, which learned grasping in simulation and could handle objects it had not seen before; Ambi Robotics now operates 80 machines sorting more than a million packages each week.
Goldberg describes research on untangling cables, folding laundry, and putting items into bags. Cable untangling reaches about 80% success, while the bagging system succeeds about half the time; a two-armed laundry robot uses fabric flinging and friction to fold more quickly.
After decades of research, robots remain much better at some tasks than at the dexterous household work people handle easily. Goldberg expects robots to become more capable, while noting that people will still need to perform many tasks machines cannot manage.
Briefs summarise talks and interviews; timestamps open the video at that moment. Claims in a talk are the speaker's own.
Bloomberg Originals host and narrator, with 1X founder Bernt Børnich, UC Berkeley professor Ken Goldberg, GXO’s Patrick Kelleher, and other robotics, industry, and labor experts, examines humanoid robots’ potential and the hurdles they face in factories, warehouses, homes, and hospitals.
Ken Goldberg, a professor at UC Berkeley, discusses how robotics can combine learned models with engineering at an IEEE Robotics and Automation Society event.
Ken Goldberg, a roboticist at UC Berkeley, presents research on robot grasping for the University of Toronto Robotics Institute, hosted by his former PhD advisor Animesh.
