π§ AI | π€ Robotics
π M. Tech in Intelligent Automation & Robotics(JU)
π Learning, building & sharing in public
π Thoughts & opinions β @iamjaminur
Hi, I'm Jaminur π
I'm currently pursuing an https://t.co/wAG6D4dHnP in IAR, and I've decided to use this account to document things that genuinely interest me, confuse me, or make me think.
Expect posts about:
π€ Robotics
π§ AI
π Research & Engineering
π Future Technologies
Korea's increased investment in Physical AI caught my attention for one reason:
The funding isn't just about flashy humanoids.
It touches areas like:
π¬ Research
π‘οΈ Defense
π₯ Healthcare
πΎ Agriculture
π€ Robotics
That's when a technology starts looking bigger than a trend.
#PhysicalAI #Robotics #AI #Korea
AI is getting better at telling a robot what to do.
But the last few millimeters are still a different story.
Recent tests of GPT-6 Astra on real robot arms reportedly showed a big improvement in gross movements, while fine manipulation remained difficult.
A model can plan the pick.
Getting the grasp exactly right is another problem.
Planning β Precision.
#AI #Robotics #PhysicalAI
China is moving extremely fast in humanoid robotics.
But there's an interesting gap:
Building robots is one challenge.
Finding useful jobs for them is another.
A robot can look incredible on a stage.
A warehouse asks different questions:
Can it work for hours?
Can it recover from errors?
Does it reduce cost?
Does it improve productivity?
Built β Deployed.
#Humanoid #Robotech #Robot
SLAM sounds simple:
"Build a map while figuring out where you are."
But underneath that sentence is a serious estimation problem.
The robot has to reason about:
π Its position
πΊοΈ The map
π‘ Sensor measurements
β Uncertainty
All while moving.
The more I study SLAM, the more I realize that autonomous navigation is really a problem of estimating the unknown.
"Currently i am working on Autonomous Path Planning Robot". This screenshot is part of my work.
#SLAM #Robotics #AutonomousSystems
Everyone is talking about humanoid robots.
But here's something I find more interesting:
The biggest robotics revolution may already be happening without humanoids.
Factories, warehouses, ports and hospitals are increasingly using robots designed for one specific job.
No face.
No human-like body.
No viral dance.
Just a machine that does its job extremely well.
Maybe the future of robotics isn't one robot that can do everything.
Maybe it's millions of specialized robots doing one thing exceptionally well.
As someone learning robotics, that's a shift I'm paying more attention to. π€
#Robotics #Automation #AI #Humanoids
AI spent the last few years living inside screens.
Now it's getting a body. π€
Cameras β perception
Sensors β awareness
AI β decisions
Motors β action
That's the idea behind Physical AI.
The interesting question isn't just:
"How smart can AI become?"
It's:
"How useful can AI become when it can actually interact with the physical world?"
As someone learning AI + robotics, this is probably one of the most exciting directions to watch right now.
#PhysicalAI #Robotics #AiGrok
Nvidia is buying Hugging Face for ~$12.9B.Thatβs not just a chip company anymore. Thatβs the storefront for open models.Good for open source β or the end of it?
#Nvidia#OpenSource