Sim2Real Deployment: RL-based Grasping Using Piper Arm - Hackster.io<br>&times;Embed the widget on your own site<br>Add the following snippet to your HTML:

Build an Isaac Lab RL grasping system for the NERO arm with PPO training, ROS2 workflows, and sim-to-real Embodied AI research.<br>Read up about this project on

Sim2Real Deployment: RL-based Grasping Using Piper Arm

AgileX Robotics

1 209

Overview<br>Things<br>Story<br>How to Build a Robot Arm RL Grasping System in Isaac Lab | NERO Arm<br>Project Summary<br>Tech Stack<br>Key Specifications<br>1. Project Setup and Environment Preparation<br>1.1 Install Isaac Lab<br>Environment<br>1.3 Clone the Repository and Install Dependencies<br>2. Environment Validation<br>3. Project File Structure<br>4. Download the URDF Model<br>5.Configuring Isaac Lab Files<br>Step 2.Create Task Configuration Files<br>6. An Isaac Lab Training and Evaluation Pipeline<br>7. Summary<br>💬 FAQ<br>Credits<br>Comments(0)

Share on FacebookShare on LinkedInShare on PinterestShare on RedditShare on XEmbed" data-html="true" data-placement="bottom" data-toggle="popover" data-trigger="click">

AgileX Robotics

Published May 20, 2026

Sim2Real Deployment: RL-based Grasping Using Piper Arm<br>Build an Isaac Lab RL grasping system for the NERO arm with PPO training, ROS2 workflows, and sim-to-real Embodied AI research.<br>BeginnerFull instructions provided209

Things used in this project

Hardware components<br>AgileX Nero&times;1Custom 3D-Printed Camera Mount&times;1Robotic Gripper&times;1RealSense Depth Camera&times;1USB 3.0 Hub&times;1Software apps and online services<br>NVIDIA Isaac LabROS 2Python 3.8+uv – Fast Python package managerConda – Virtual environment management

Story

How to Build a Robot Arm RL Grasping System in Isaac Lab | NERO Arm<br>This project presents a reinforcement learning workflow for Embodied AI manipulation built on the Nero robotic arm, SO-ARM101, and NVIDIA Isaac Lab . It establishes a simulation-driven framework for training and evaluating robotic manipulation policies, with a focus on preparing the system for simulation-to-real transfer.

Project Summary

Tech Stack<br>RL training pipeline<br>policy validation process<br>robotic manipulation task configuration<br>simulation-to-real transfer preparation

Key Specifications<br>Programming Language : Python 3.8+<br>Hardware : Nero Robotic Arm https://global.agilex.ai/products/nero<br>Base Framework : SO-ARM101 https://github.com/MuammerBay/isaac_so_arm101<br>Simulation Platform : NVIDIA Isaac Lab<br>Open-source implementation :https://github.com/agilexrobotics/Agilex-College/tree/master/isaac_sim/agx_arm_IsaacLab

1. Project Setup and Environment Preparation

1.1 Install Isaac Lab<br>Follow the official guide to install Isaac Lab:<br>👉 Isaac Lab Pip Installation Guide<br>We use the pip-based installation method (recommended).

Environment:<br>Conda virtual environment<br>Python development environment<br>NVIDIA Isaac Lab<br>Nero robotic arm project dependencies

1.2 Install the uv Package Manager<br>This project uses uv as its Python package manager.<br>As a fast, next-generation tool, uv delivers:<br>Faster package installation<br>Efficient dependency resolution<br>Built-in virtual environment management<br>Compared to traditional tools like pip, uv streamlines setup and reduces environment issues in Python-based robotics and embodied AI workflows.<br>First, install uv with a single command:<br>curl -LsSf https://astral.sh/uv/install.sh | shAfter installation, restart your terminal or run the following command to activate the uv environment:<br>source $HOME/.cargo/env

1.3 Clone the Repository and Install Dependencies<br>Next, clone the project repository, enter the project directory, and use `uv` to install all required dependencies with one command:<br>git clone https://github.com/smalleha/isaac_so_arm101.git<br>cd isaac_so_arm101<br>uv syncuv will automatically create a virtual environment and install all necessary dependency packages. The entire process usually takes only a few minutes and is significantly faster than traditional pip-based installation workflows.

2. Environment Validation<br>To validate the setup for tasks, we first verify that the required simulation environments for the Nero robotic arm and Piper are properly registered:<br>uv run list_envsThe expected output should include Isaac-Nero-Reach-v0 and Isaac-Piper-Reach-v0, confirming that the environments have been installed successfully.<br>Next, run a simulation test with a zero-action agent to validate environment execution and ensure the robotic control pipeline works as expected:<br># Test the Piper environment with a zero-action command<br>uv run zero_agent --task Isaac-SO-ARM100-Reach-v0If the simulation window launches and the robotic arm behaves as intended, the environment is confirmed to be ready.

3. Project File Structure

This directory structure provides a clear overview of the project organization, making it easy to extend with new use cases such as Nero robotic arm example.

4. Download the URDF Model<br>This project uses the Nero URDF model from the agx_arm_urdf repository. After cloning the repository,<br>copy...