Summary of Why Don’t We Have Better Robots Yet? | Ken Goldberg | TED

In the TED talk "Why Don’t We Have Better Robots Yet?" by Ken Goldberg, the speaker discusses the current state of robotics, the challenges faced in developing more capable robots, and the gap between the capabilities of robots and human abilities.

Key Technological Concepts and Product Features:

• Moravec's Paradox: This principle highlights that tasks easy for humans, like grasping and manipulating objects, are challenging for robots, while tasks that are difficult for humans, like lifting heavy objects, are easier for robots.
• Robot Grippers:
  • Complex Robot Hands: These are heavy, expensive, and often unreliable.
  • Simple Grippers: Examples include parallel jaw grippers and suction cups, which are lightweight, reliable, and cost-effective. Simple designs can achieve complex tasks, as demonstrated by human-operated simple grippers.
• Sensor Technologies:
  • High-Resolution Cameras: Used for capturing images but lack depth perception.
  • LIDAR: A breakthrough technology that builds 3D models of environments using light beams, though it struggles with shiny or transparent objects.
  • Tactile Sensors: Still in early development, these sensors use cameras to analyze surfaces during contact.
• Uncertainty in Robotics: Challenges arise from uncertainties in control, perception, and physics, making precise manipulation difficult.
• E-commerce Robotics: The demand for robots in warehouses has increased, especially post-pandemic. The company Ambi Robotics, founded by Goldberg and his students, utilizes AI and deep learning to develop robots capable of reliably picking and sorting packages.
• Research Projects:
  • Untangling Knots: Robots are being trained to manipulate cables and untangle them, achieving about 80% success.
  • Laundry Folding: New methods aim to increase the speed of laundry folding robots, utilizing techniques like the "two-second fold."
  • Bagging: Robots have been trained to manipulate bags using fluorescent paint for better recognition, achieving success in about half of the attempts.

Conclusion:

Goldberg emphasizes that while progress is being made in robotics, significant challenges remain. Robots are not yet capable of performing many tasks that humans find easy. He encourages patience as advancements continue to be made in the field.

Main Speaker:

• Ken Goldberg, Professor at UC Berkeley and researcher in robotics.

Notable Quotes

— 00:04 — « I have a feeling most people in this room would like to have a robot at home. »

— 00:26 — « I mean, OK, wait, we do have some robots in our home, but, not really doing anything that exciting, OK? »

— 01:10 — « What's easy for robots, like being able to pick up a large object, is hard for humans. But what's easy for humans, like being able to pick up some blocks and stack them, well, it turns out that is very hard for robots. »

— 11:14 — « What's easy for robots is hard for humans. And what's easy for us is still hard for robots. »

— 12:00 — « Because we want the robots, but robots also need us to do the many things that robots still can't do. »

Category

Technology

Video