Marc Raibert's Legacy: Pioneering Dynamic Legged Robotics and the Future of Lifelike Machines

Marc Raibert, a legendary roboticist and founder of Boston Dynamics, shares his lifelong journey from a builder inspired by his engineer father to a pioneer in dynamic, legged robotics. His initial fascination with a disassembled robot arm at MIT in 1974, coupled with a disillusionment with neurophysiology's conceptual limitations, propelled him into the nascent field of AI and robotics. This foundational experience at MIT's AI lab, and later the Leg Lab at Carnegie Mellon, set the stage for his distinctive approach to robot design, emphasizing functionality over static aesthetics and drawing inspiration from biological movement.

Raibert's core philosophy diverged sharply from the cautious, tripod-stable robots prevalent at the time. He advocated for an "aggressive" approach, inspired by how humans and animals utilize bouncing, springiness, and predictive balance in locomotion. This led to the development of the groundbreaking one-legged hopping robot, a "pogo stick robot," which initially focused on energy management before evolving to tackle complex 3D balancing. He highlights the community's tendency towards "safe" static grasping in manipulation, arguing for a more dynamic, fumbling-tolerant approach akin to human dexterity, citing Julia Child's diverse handling techniques as an example.

The development of the hopping robot was a testament to early collaboration and unconventional funding. Ivan Sutherland, a computer graphics pioneer, provided initial seed money, leading to a pivotal meeting with DARPA's Craig Fields, who saw the potential in Raibert's skeletal prototype and provided significant funding. Raibert details the technical challenges of controlling a 3D hopping robot, involving energy regulation, precise foot placement relative to the center of mass, and maintaining body attitude. He emphasizes that early solutions were often "sketches" that later generations optimized, and credits collaborators like Ben Brown for making these ambitious ideas a reality, despite initial skepticism from peers who considered legged locomotion a "waste of time."

Boston Dynamics, founded in 1992, initially focused on physics-based simulation before a project to make Sony's AIBO run faster reignited their passion for physical robots. A significant pivot involved abandoning a technically impressive but commercially unviable surgical simulator with force feedback, solidifying the company's commitment to robotics. The creation of Big Dog, funded by DARPA's biotics program and propelled by engineers like Martin Buehler, marked a major milestone, establishing Boston Dynamics' reputation for robust, real-world capable machines. Raibert reflects on the public's anthropomorphic connection to robots like Spot and Atlas, noting the contrast between media portrayals of fear and the public's actual desire to interact with and even "pet" the robots, underscoring the profound human-robot relationship that has emerged from their relentless "build it, break it, fix it" ethos.

people who think you don't need to innovate Hardware anymore are wrong

I just looked at that and said wow that's wrong you know that's not that's not anything like how people and animals work because we bounce and fly

I think it's all too safe and I think trying to break out of that safety thing of static grasping

maybe flaws is a good thing Perfection is not necessarily the right thing to be chasing

probably the smartest thing I ever did is to find the other people

I don't know if they doubted whether it was possible but I think they thought it was a waste of time

the media likes to say that they're terrifying and and that people are afraid and and YouTube commenters like to say that it's frightening but when you take a spot out there now maybe it's self-selecting but you get a crowd of people who want to take pictures want to pose for selfies want to operate the robot want to pet it want to put clothes on it

we never had investors until Google bought us which was after 20 years

we killed it and we moved on and that was really a milestone in the company because we you know we sort of understood who we were and uh and what would work and what wouldn't

you know as an academic I was mostly just trying to do the next thing you know make some progress impress my colleagues if I could and have fun

build it, break it, fix it