Russ Tedrake is a roboticist and professor at MIT and vice president of robotics research at TRI. He works on control of robots in interesting, complicated, underactuated, stochastic, difficult to model situations. Support this podcast by supporting our sponsors. Click links, get discount: - Magic Spoon: https://magicspoon.com/lex & use code LEX at checkout - BetterHelp: https://betterhelp.com/lex - ExpressVPN: https://www.expressvpn.com/lexpod If you would like to get more information about this podcast go to https://lexfridman.com/ai or connect with @lexfridman on Twitter, LinkedIn, Facebook, Medium, or YouTube where you can watch the video versions of these conversations. If you enjoy the podcast, please rate it 5 stars on Apple Podcasts, follow on Spotify, or support it on Patreon. Here's the outline of the episode. On some podcast players you should be able to click the timestamp to jump to that time. OUTLINE: 00:00 - Introduction 04:29 - Passive dynamic walking 09:40 - Animal movement 13:34 - Control vs Dynamics 15:49 - Bipedal walking 20:56 - Running barefoot 33:01 - Think rigorously with machine learning 44:05 - DARPA Robotics Challenge 1:07:14 - When will a robot become UFC champion 1:18:32 - Black Mirror Robot Dog 1:34:01 - Robot control 1:47:00 - Simulating robots 2:00:33 - Home robotics 2:03:40 - Soft robotics 2:07:25 - Underactuated robotics 2:20:42 - Touch 2:28:55 - Book recommendations 2:40:08 - Advice to young people 2:44:20 - Meaning of life
Russ Tedrake shares about his experience leading MIT's DARPA robotics challenge team, as well as elite barefoot running and the control of under-actuated dynamical systems like the human body.
The passive dynamic walker, which takes design inspiration from a rimless wheel, showcases a more natural and graceful human-like motion in robots compared to traditional walking robots.
Despite having no brain, motors, or control, dead fish can swim upstream in vortex streets due to the mechanics of their bodies resonating with the flow. This phenomenon baffles scientists and challenges our understanding of locomotion.
The course of history and evolution can turn on seemingly insignificant events. For instance, the evolution of humans walking on two feet might have been influenced by the need for communal food gathering. The conversation also touches on the impact of the environment and the benefits of a meat-only diet.
The podcast discusses the tension between focusing on long-term goals and personal habits, such as exercising, and the importance of balancing the two.
In this episode, the importance of rigorous thinking in understanding complex systems is discussed. The guest emphasizes the need to take a class on analysis to develop a taste for rigorous thinking.
MIT competed in the DARPA Robotics Challenge where teams could enter as a hardware team that brought their own robot or as a software team that would try to win the right to use one of the Boston Dynamics robots. The challenge aimed to address degraded networks in natural disasters and was part of a series of challenges organized by DARPA, including the Grand Challenge and Urban Challenge that focused on self-driving cars.
The speaker reflects on the hard work of robotics students and how the optimization problem they are solving doesn't change much from one time step to the next, which is beneficial for his controller's performance.
The hosts discuss the challenges faced when creating a semi-autonomous system and the importance of integration testing.
The idea of robots that can win in a wrestling match or fight against a human may seem like a far-off fantasy, but researchers are developing technology to improve robot capabilities in physical challenges. At an annual robotics competition, a team faced unforeseen challenges with their robot, forcing them to rely on a thorough checklist to ensure the robot's best performance.
RoboCup has a serious following with a rich history, whereas the Pick Three Game, created by Josh Tenenbaum, is fairly new. Winning the game of soccer as humans play it would require robots to play very differently, while techniques in the Pick Three Game are learned more intellectually and instinctually.
The public's perception of animal-like robots can be misconstrued as having a level of self-awareness or consciousness, leading to fears of a surveillance state. However, this perception may be a result of cultural influences such as movies like Terminator.
Controlling a Jibo robot requires a different approach compared to an autonomous car or drone. Humans have the ability to create simple models quickly, which can aid in the control of robots through familiar models of human and animal behavior.
Drake is a modeling language for complex systems that requires a little more description about the system than traditional simulators. It is not just a simulator, but a platform where controllers, perception systems, low-level control and even planning and optimization can be written.
Researchers are exploring the hard corner cases in manipulation and using problems like loading the dishwasher to study those scenarios. Their goal is to develop algorithms to create a commercial product that can perform the task with confidence for any dish in anyone's kitchen.
The discussion revolves around whether or not robots will become more human-like with actuators that look like arms in homes in the future, or if they will maintain their current form, similar to a dishwasher. Fleet learning and its possibilities for improving robots is also discussed.
Optimal control and numerical optimal control are used to control underactuated systems. To control dynamic robots or airplanes, it is necessary to embrace underactuated dynamics.
The use of deep learning systems to analyze physical systems such as onions or cars produces a high-dimensional state representation that is relevant for understanding the state of the system.
While robots have become increasingly prevalent in our society, they still lack the human element that is necessary for a good experience for the worker. By adding touch and making robots look less intimidating, it could completely change the way people perceive and interact with them.
The dialogue that occurs within a good book is something that cannot be fully captured in the written word. Even with the shift to virtual lectures and seminars, the passion and realization of the value of good literature remains the same.
Deep thinking is crucial for success in various aspects of life, including robotics, career, relationships, and personal growth. Those who can think critically are more likely to become thought leaders and excel in their fields.
In this podcast, Lex Friedman talks about the significance of stepping out of our comfort zone and the importance of exploring and learning different things. He also shares the words of Neil deGrasse Tyson about robots in space and human-based space exploration.