I really enjoyed this conversation with Vijay. Here's the outline: 0:00 - Introduction 0:58 - First robot 3:37 - Proudest accomplishments 5:32 - Drone, UAV, aerial robot terminology 6:23 - Biologically inspired robotics 8:34 - Swarm as an individual organism 11:18 - Distributed control 15:04 - Types of flying robots 19:57 - Math in a TED talk 20:29 - What does it take to make a robot fly? 27:09 - Getting from point A to point B 29:22 - Machine learning in robotics 33:53 - Autonomous vehicles 37:05 - Autonomous driving vs autonomous flight 38:52 - Applications of robot swarms 40:12 - Batteries 42:30 - Flying cars 44:03 - Robots and humans 49:48 - Sci-fi inspired fears of robots 52:22 - Open problems in robotics
So many postive comments for this and for many of Lex's interviews. I agree with all of the positive comments. Ah, to be able to snap my fingers and to find that I am in my late teens and in college again. That is the desire that listening to these interviews ignites in me. 🌻
Lex, so far, in my experience, you are the only person on RU-vid whose videos are so carefully organized and documented in terms of discussion matter and time. Especially for such advanced topics. You can simply select a defined point in the video that you want to hear or re-listen to without having to manually scroll through and waste time. Your extra hard work doing that for us makes watching your videos an absolute pleasure to listen to. Thank you and keep it up.
Thanks Lex! Very interesting conversation that gives general audience a sense of where we are at for robotic technology. Certainly will inspire the next generation of engineers!
Lex: just started watching ur interviews...They are great! I could say more but it is true that simplicity is beautiful thing. Vijay seems like such a nice guy and very intelligent; aigh, and patient...
listening to Vijay talk about the 'energy problem'. It reminds me of the human problem with satisfying energy requirements for our own brain, using tools and finding better ways to gather/trap/harvest food. (Homo Erectus, ~1.8 million years ago)
17:30, 20:00 agility real means and mathematics giving goosebumps still watching lets see how much more goosebumps coming up, 33: 00 is he talking about navigation and action in outdoor, unstructured , dynamical environment. the actual problem in robotics, in last 52:22 open problems , the importance of representation .... now i may not be able to sleep for few weeks properly... all i got to know that my approach is on right track.....
Vijay is an interesting dude and his comment on designing at a 'higher dimensional whole' should be implemented in all AGI design in particular in artificial cognition to capture amazing advantages. It also highlights how higher levels of human co-operation (as a whole) across all types of humans would lead to a greater AI whole. The other option is to hide safely behind our intellectual gates and avoid sharing with those on the outside. Time is finite, I prefer the former.
Here's the idea for home delivery ... Raise the packages in a blimp in the morning, deliver the packages with a mostly gliding robot that drives back to the blimp for the next day.
My question is why after half a century hasnt this shown up in everyday life? it seems like its been funded heavily but like many techs it seems slow to get out to where everyone can afford it . phones took 25 30 years maybe .
"maybe we can elect some engineers to office as well" With all due respect, Lex, before making that wish I think you may want to study some economics, public choice economics in particular, and perhaps some history, e.g., Crisis and Leviathan, which I think is appropriate at this time.
35:55 Well what's human performance on that safety critical task? Answer: not 99.9%. So why subject the machine learning system to that kinda requirement
Time wasting.when we alreadyhad antigravity.technology available.please Google search emreySmith and cooreygoodey..these two people's can give you the correctanswer.