UC Berkeley's Ken Goldberg Answers Your Top Robotics Questions

Subscribe here for future updates

As the robotics industry continues to evolve, we’re excited to bring you a series of interviews with some of the top minds in robotics. In this first installment, we’re joined by Dr. Ken Goldberg, a professor and the William S. Floyd Jr. Distinguished Chair in Engineering at UC Berkeley, co-founder and chief scientist at robotics parcel sorting startup Ambidextrous, and a fellow at IEEE.

Table of Contents

Generative AI in Robotics: A Game-Changer?

Dr. Goldberg’s first topic of discussion is the impact of generative AI on the field of robotics. "Although the rumblings started a bit earlier," he explains, "2023 will be remembered as the year when generative AI transformed robotics."

Large language models like ChatGPT have enabled robots and humans to communicate in natural language, opening up new possibilities for collaboration and interaction. Words have evolved over time to represent useful concepts from ‘chair’ to ‘chocolate’ to ‘charisma.’

But that’s not all – roboticists have also discovered that large Vision-Language-Action models can be trained to facilitate robot perception and control the motions of robot arms and legs. Training requires vast amounts of data, so labs around the world are now collaborating to share data.

The Future is Multi-Modal

Another exciting topic in robotics is the development of ‘Multi-Modal’ models. In this context, Dr. Goldberg explains that multi-modal refers to combining different input modes, such as vision and language, to enable robots to better understand their environment.

But what’s even more exciting is the potential for these models to be applied in real-world scenarios, such as robot-assisted surgery or search and rescue operations.

A Fundamental Problem: Singularities

Dr. Goldberg also touches on a fundamental problem in robotics – singularities. "For repetitive robot motions," he explains, "singularities can be avoided by tedious manual fine-tuning of repetitive robot motions to adjust them such that they never encounter singularities."

However, for the growing generation of applications where robot motions are not repetitive, including palletizing, bin-picking, order fulfillment and package sorting, singularities are common. They are a well-known and fundamental problem as they disrupt robot operations at unpredictable times (often several times per hour).

A New Solution: Jacobi Robotics

To address this issue, Dr. Goldberg co-founded a new startup, Jacobi Robotics, which implements efficient algorithms that are guaranteed to avoid singularities. This can significantly increase reliability and productivity for all robots.

The Future of Robotics: A Conversation with Ken Goldberg

In the coming weeks, we’ll be publishing more interviews with top robotics experts. Stay tuned for: