TossingBot: Learning to Throw Arbitrary Objects Researchers have trained a robot to throw objects using visual observations and trial and error, overcoming challenges posed by varied object poses and properties. By combining simulation and deep learning, the system achieves accuracy and generalization for new objects and target locations. This technology might find applications in sorting rubbish or kitting non-sensitive parts, allowing robots to rapidly place objects beyond their kinematic range in pick-and-place scenarios. Thoughts ⁉️ #robotics #simulation #gripper #research #cobots Via: Andy Zeng from Princeton University Inspired by: Daniel Kuepper
Can it do it with a basketball players make way to much money and all the injuries we pay players to sit on the bench
Gives new meaning to ‘throwing something over the wall’. This is one way to speed up the pick cycle but seems highly dependent on the catch and the material properties of the items being tossed.
Different functionality usecase for sorting of packages for the good 😊
Just teach all like back in the sandbox. Users ánd Robots. That's how normal life is also trained. Grow up! Together!
Very impressive! I would be interested to know how does the robot determine the weight of the object it is grabbing and how is the release time for throwing determined?
Long story short you may have named your robot the equivalent of Jerkingbot (for the UK people.) FYI.
Yeah, they might want to consider re-naming it for the UK Market...
I wander what kind of robots these are
Hey-can I use this next time at the Arcade?
🇨🇭 𝗪𝗼𝗿𝗹𝗱-𝘄𝗶𝗱𝗲 #𝟭 𝗥𝗼𝗯𝗼𝘁𝗶𝗰𝘀 𝗩𝗼𝗶𝗰𝗲 on LinkedIn 🤖 | 𝗔𝗡𝗬𝗯𝗼𝘁𝗶𝗰𝘀 𝗖𝗖𝗢 | CRO, CSO, CMO | Happy dad of 👧🏼👶🏼 | 𝑭𝒐𝒍𝒍𝒐𝒘 𝒎𝒆 🔔 to stay updated on #robotics and #industrialautomation
9moIntereting video with great learning David