Robotics and humanoid robotics & civilian drones discussion
- Thread starter tphuang
- Start date
A retrospection on humanoid robotics industry.
OpenAI failed miserably in robotics, shuttering their Dactyl team. They've since then partnered with experts in Figure and poached talent from Meta Reality Lab. I wonder if any competitors to Chinese firms will actually come out of it.
OpenAI failed miserably in robotics, shuttering their Dactyl team. They've since then partnered with experts in Figure and poached talent from Meta Reality Lab. I wonder if any competitors to Chinese firms will actually come out of it.
Mechanical part of common robot seems to be more or less developed. Now is the time for software to catch up, so robots can do actions without too much human input and detailed commands.
I wonder if initial household robots will first need weeks of calibration and observing workflow examples from their owners.?
I wonder if initial household robots will first need weeks of calibration and observing workflow examples from their owners.?
I'm still a bit sceptical. Similar to self driving cars the first 80% is easy, the last 20% seems insurmountable.
Household robots seem an entire level above cars due to no two homes being the same, you cannot simply train a robot at the factory for all homes.
Unitree believes it can lower robot dog price to 3000 to 4000 RMB, that's lower than iPhone price
I doubt they’ll have automation or AI at that price though. Still, this is impressive even on pure hardware level.
Sure it's hard. That's why didn't dare give a timetable.
And I agree you can't just train a robot at a facility. That is why I mentioned calibration at user's home and observing the desired work flow. In a way I expect users will have to aid the programming of their household robots. But not through inputting code. Rather, by showing and telling robots what they did wrong and how to do it next time.
Like having a very obedient little child.
Stair climbing robot