This project provides a complete set of tools for robot learning. It helps you train robots to perform tasks through data collection, skill labeling, and reinforcement learning. You can use this software to teach robots how to move, react to their environment, and fix mistakes through human guidance.
Before you install this software, ensure your computer meets these requirements:
- Operating System: Windows 10 or Windows 11.
- Processor: Intel Core i7 or AMD Ryzen 7 (or better).
- Memory: 16 GB RAM (32 GB recommended).
- Storage: 50 GB of free space on a solid-state drive (SSD).
- Graphics: An NVIDIA GPU with at least 8 GB of video memory is required for training tasks.
You can download the latest version of the application from our release page.
Once you reach the page, look for the section labeled "Assets." Click the file that ends in .exe to start your download. Save this file to a folder where you can find it easily, such as your Downloads folder or a dedicated project folder on your desktop.
- Run the installer: Double-click the file you downloaded. Follow the prompts on the screen to finish the installation.
- Verify dependencies: The installer will check if your system has the necessary drivers for your graphics card. If it detects missing components, it will provide a link to download the required software from the manufacturer website.
- Launch the program: Once the installation finishes, you will see a shortcut icon on your desktop. Double-click this icon to open the application.
This software operates on a closed-loop system. This means the robot learns by doing, checking its work, and adjusting based on feedback.
- Data Collection: You record the robot performing tasks. This creates the base data.
- Labeling: You mark the progress of the robot. You highlight when the robot succeeds and when it fails.
- Training: The software processes your data to create a strategy. It learns which actions lead to success.
- Deployment: The robot uses the strategy to perform tasks in the real world.
- Correction: If the robot makes a mistake, you guide it through the correct action. The software uses this new data to improve the next version of the strategy.
The software divides tasks into several parts so you can work on them one by one.
- Model Training: This part handles the brain of the robot. It turns your recorded data into a set of instructions for the machine.
- Progress Tracking: This tool measures how close the robot is to finishing a task. It uses visual data to guess the probability of success.
- Feedback Loop: This section manages the information you provide when you correct the robot. It filters this data to ensure the training process focuses on your specific instructions.
- Hardware Bridge: This part connects your computer to the physical robot. It translates your digital strategy into signals the robot hardware understands.
- Open the application.
- Select "Create New Project" from the main menu.
- Choose a name for your project and pick a folder on your computer to save your data.
- Connect your robot to the computer using the cable provided by the robot manufacturer.
- Navigate to the "Input" tab to start recording your first session.
- Perform the task you want the robot to learn. Keep your movements smooth and clear.
- Stop the recording when you finish the task.
- Go to the "Training" tab and click "Start Training." The software will now process the data you just created.
- Watch the progress bar to monitor the speed of the training. Depending on the complexity of your task, this might take several minutes or longer.
- The software does not launch: Check if your graphics drivers are up to date. Visit the NVIDIA website to download the latest drivers for your specific card.
- Training is too slow: Ensure you have enough free space on your SSD. Move the project folder to a drive with faster read and write speeds.
- Robot does not respond: Check the physical connection between your robot and the computer. Ensure the hardware is turned on and in the correct mode for remote control.
- Data errors: If the software reports an error during training, check your recordings. Ensure your video files are clear and that the robot is visible at all times.
Keywords: robot learning, artificial intelligence, robotics, machine learning, windows application, data collection