unitree_lerobot

unitreerobotics/unitree_lerobot is ranked 13,926 among these 8 peers; ikergarcia1996/Self-Driving-Car-in-Video-Games is ranked 13,924.

📖 Introduction This repository is used for (Supports LeRobot datasets version 2. 1 🦾 LeRobot Environment Setup The purpose of this project is to use the LeRobot open-source framework to train and test data collected from Unitree robots. Therefore, it is necessary to install the LeRobot-related dependencies first. The installation steps are as follows, and you can also refer to the official LeRobot installation guide: 1. 2 🕹️ unitree_sdk2_python For on Unitree robots, some dependencies need to be installed

. 1 🖼️ Load Datasets If you want to directly load the dataset we have already recorded, Load the dataset from Hugging Face. If you want to load data from a local source, please change the parameter. 2 🔨 Data Collection If you want to record your own dataset. The open-source teleoperation project avp_teleoperate can be used to collect data using the Unitree G1 humanoid robot. For more details, please refer to the avp_teleoperate project. 3 ✂️Data Processing When you

finished the Data Collection step, it may be essential for you to process the data you have been collected. For example, if one of the episodes failed to complete the task, you need to delete the episode, to increase the data's quality. We have designed a data editor in this project, in order to help users process the data. Currently, it can cut off excess segments in the episodes, and delete the bad episodes. Before you launch the data editor

, you need to install pyqt5. At first, click the button , choose the path of your dataset. For example, at the path tree below, is the path you need to choose. You can drag the red line directly to seek playback, and drag the red line with Shift on your keyboard to select the playback range, and then trim the selected range with the button . If you want to delete the current episode, click the button . 4 🛠️ Data Conversion The data

collected using avp_teleoperate is stored in JSON format. Assuming the collected data is stored in the , the format is as follows 2. 1 🔀 Sort and Rename When generating datasets for LeRobot, it is recommended to ensure that the data naming convention, starting from , is sequential and continuous. You can use the following script to the data accordingly. 2 🔄 Conversion Convert Dataset to Format. You can define your own based on ROBOT_CONFIGS. **Node:** is a robot type used for

data collection in unitree_sim_isaaclab , with the head equipped with a single-viewpoint camera. 🚀 Training For training, please refer to the official LeRobot training example and parameters for further guidance. Please refer to it in detail Please refer to it in detail Please refer to it in detail Please refer to it in detail Please refer to it in detail If you want to use multi-GPU training, please refer to the details here 4. 🤖 Real-World Testing To

test your trained model on a real robot, you can use the eval_g1. py script located in the eval