Physical AI Vision PnP Simulator

Physical AI Enabled Vision Pick-and-Place Simulator 🤖

Traditional robotic pick-and-place systems require engineers to manually write motion programs whenever object arrangements change. This project explores a simpler workflow where users describe a desired arrangement in natural language and allow an AI agent to generate the required robot motion plan automatically.

This project demonstrates a complete Physical AI workflow that combines Computer Vision, Camera & Workspace Calibration, AI-based Task Planning, and RoboDK Simulation to enable natural-language control of a UR5e robot.

Using simple commands such as "Stack the red block on the blue block," the system detects objects, computes real-world coordinates, generates robot motion plans with AI, and executes them in simulation automatically.

🔹 Vision-Based Object Detection
🔹 Homography-Based Coordinate Mapping
🔹 Natural Language Robot Programming
🔹 AI Motion Planning
🔹 RoboDK Simulation
🔹 UR5e Pick-and-Place Automation

Source Code:
https://github.com/dhanushmaverick/Python-Physical-AI-Project

Developed by:
Dhanush Vulli Bala
Praneeth Manickam Srinivas
Joel George Thomas
ignore hashtags:
#PhysicalAI #Robotics #ArtificialIntelligence #ComputerVision #RobotLearning #RobotProgramming #Automation #IndustrialAutomation #MachineLearning #RoboticsEngineering #UR5e #UniversalRobots #RoboDK #PickAndPlace #RobotSimulation #AIAgent #VisionAI #SLAM #PhysicalIntelligence #EmbodiedAI #OpenAI #EngineeringProjects #PythonProgramming #Mechatronics #AutonomousSystems #SmartManufacturing #FutureOfRobotics #Engineering #TechInnovation #AI

Видео Physical AI Vision PnP Simulator канала DHANUSH