Robots Doing Stuff #44 - Placing Object in Plastic Bag
HTML-код
- Опубликовано: 8 июн 2023
- Grabbing an object as thin as a plastic bag is an incredible challenge for robotic hands, but Phoenix robots have the fine manipulation required to do it.
Learn more at sanctuary.ai/
#RobotsDoingStuff #PhoenixRobot #Humanoid #GeneralPurposeRobot #TeleOperation 
Наука
I appreciate the explanation of the chest light color meanings on the podcast
Happen to remember which color meant what mode? lol. I forgot and don't wanna try to sift through the vid again even though I enjoyed it.
Purple is analogous teleoperation (a person in a rig is remotely controlling the robot).
This is the way
Impressive!
Nice!
Why no fully autonomous demos?
We are working on a fundamentally new approach to the AGI problem where we are connecting language (verbs, nouns, adverbs, adjectives) to actions in the real world. This ground-up approach requires solving a lot of very hard problems! When we have stuff to show that's impressive enough we will start adding those to the robots doing stuff series.
@Geordie Rose Thanks. Looking forward to seeing it in future.
@@geordierose3242 language to actions? So people saying something and AI doing something as a result?
@@Srindal4657 Not exactly! We naturally connect our experience of the world using language, but machines don't -- you have to build those connections. For example, if I ask the robot to assemble a widget and show it how to do that, we are building a system where the robot can understand what is being asked by mapping the verbs, nouns, adjectives, and adverbs used by the person it is communicating with to sequences of specific instructions that it can perform. This connection between natural language and 'robot language' is one way of defining the AGI problem.
Can you please start putting in the description whether or not the robot is being piloted or if it is acting in it's own capacity! I can't help but wonder this every video I see. It sounds like from some of @geordierose3242 comments below that perhaps there is a lot more to these videos than what is being explained. I'd love to know more. I wish there was a bit more to the description or maybe even a simple voice over giving us more context.