Over the next few weeks, TechCrunch’s robotics newsletter, Actuator, will be conducting Q&As with some of the best minds in robotics. Subscribe here for future updates.
Part 1: CMU’s Matthew Johnson Roberson
Part 2: Max Bajracharya and Ross Tedrick of the Toyota Research Institute
Part 3: Meeta Dhruv Patra
This time, it’s Boston Dynamics’ CTO, Aaron Saunders. He has been working with the company for more than 20 years, most recently as Vice President of Engineering Affairs.
What role(s) will generative AI play in the future of robotics?
The current rate of change makes it difficult to predict the distant future. Foundational models represent a major shift in how we build the best machine learning models, and we’re already seeing some impressive near-term acceleration in natural language interfaces. They provide opportunities to create conversational interfaces for our robots, improve the quality of existing computer vision functions, and perhaps enable new customer-facing capabilities such as visual question answering. Ultimately, we feel that these more scalable architectures and training strategies are likely to extend beyond language and vision to automated planning and control. Being able to interpret the world surrounding a robot will lead to a richer understanding of how to interact with it. It’s a really exciting time to be a roboticist!
What are your thoughts on the humanoid form factor?
Humanoids are not necessarily the best form factor for all tasks. Take Stretch, for example – we originally sparked interest in the box-moving robot from a video we shared about Atlas moving boxes. Just because humans can move boxes doesn’t mean we’re the best form factor to complete that task, and we’ve ultimately designed a custom robot in Stretch that can move boxes more efficiently and effectively than a human. However, we see great potential in the long-term pursuit of general-purpose robots, and the human form factor is the most obvious match for a world built around our form. We have always been excited about the potential of humanoid robots and are working hard to bridge the technology gap.
After manufacturing and warehouses, what is the next major category of robots?
These two industries still stand out when you look at matching customer needs with the latest in technology. As we spread, I believe we will slowly move from environments characterized by determinism to those characterized by higher levels of uncertainty. Once we see widespread adoption in automation-friendly industries like manufacturing and logistics, the next wave will likely occur in areas like construction and healthcare. Such sectors represent compelling opportunities because they have a large workforce and high demand for skilled labor, but the supply does not meet the need. Combine this with business environments, which lie between a highly regulated industrial environment and a completely unregulated consumer market, this could represent a natural next step on the path towards the overall goal.
How real are general purpose robots?
There are many difficult problems that stand between today’s general-purpose robots. Purpose-built robots have become a commodity in the world of industrial automation, but we are now seeing the emergence of multi-purpose robots. For robots to be truly general-purpose, they will need to navigate unstructured environments and address problems they have not encountered. They will have to do this in a way that builds trust and delights the user. They will need to offer this value at a competitive price. The good news is that we are seeing a dramatic increase in critical mass and interest in this field. Our children are exposed to robotics early, and recent graduates are helping us achieve a tremendous acceleration in technology. Today’s challenge of delivering value to industrial customers paves the way toward future consumer opportunities and the general purpose future we all dream of.
Will home robots (other than vacuum cleaners) take off in the next decade?
We may see more robots introduced into the home in the next decade, but for very limited and specific tasks (like the Roomba, we will find other instances of clear value in our daily lives). We are still more than a decade away from achieving multi-functional home robots that provide value to the broad consumer market. When will you pay the same amount for a robot as you pay for a car? When you achieve the same level of reliability and value you’ve come to take for granted in the amazing machines we use to move us around the world.
What’s an important robotics story/trend that doesn’t get enough coverage?
There is a lot of excitement about artificial intelligence and its ability to change all industries, including robotics. Although it has a clear role and may open up areas that have been relatively static for decades, there is much more to a good automated product than just 1’s and 0’s. For AI to achieve the physical embodiment it needs to interact with the world around us, we need to track advances in key technologies like computers, perceptual sensors, power supplies, and all the other parts that make up a complete robotic system. The automotive sector’s recent shift towards electrification and advanced driver assistance systems (ADAS) is rapidly transforming a huge supply chain. Advances in graphics cards, computers, and increasingly sophisticated AI-enabled consumer electronics continue to drive value into adjacent supply chains. This massive snowball, which is rarely in the spotlight, is one of the most exciting trends in robotics because it enables innovative small companies to stand on the shoulders of giant corporations to create exciting new products.