darshmenon/pickplace-rl-mobile-manipulator

Autonomous mobile manipulator built with ROS 2 Jazzy and Gazebo Harmonic, integrating SAC-based reinforcement learning, RGB-D perception, LiDAR navigation (Nav2), safety supervision, and domain randomization for sim-to-real pick-and-place tasks

24
/ 100
Experimental

This project creates an autonomous robotic system that can learn to pick up and place objects without human-coded instructions. It takes in live camera data and object locations, and outputs robot arm and base movements to complete the task. This system is designed for robotics engineers or researchers working on automated material handling, flexible manufacturing, or autonomous warehousing applications.

Use this if you need a mobile robot to perform pick-and-place tasks autonomously, where objects might be in varying locations and you want the robot to learn complex manipulation skills directly through experience.

Not ideal if your application involves high-precision, fixed-path manufacturing or if you require human operators to program exact robot movements.

mobile-robotics autonomous-manipulation material-handling robot-learning flexible-manufacturing
No License No Package No Dependents
Maintenance 13 / 25
Adoption 4 / 25
Maturity 7 / 25
Community 0 / 25

How are scores calculated?

Stars

8

Forks

Language

C++

License

Last pushed

Mar 20, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/darshmenon/pickplace-rl-mobile-manipulator"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.