9 posts tagged with "robotics"

Working with real-world robot data depends on how ROS (Robot Operating System) messages are stored. In the article 3 Ways to Store ROS Topics, we explored several approaches — including storing compressed Rosbag files in time-series storage and storing topics as separate records.

In this tutorial, we'll focus on the most common format: .bag files recorded with Rosbag. These files contain valuable data on how a robot interacts with the world — such as odometry, camera frames, LiDAR, or IMU readings — and provide the foundation for analyzing the robot's behavior.

You’ll learn how to:

• Extract motion data from .bag files
• Create basic velocity features
• Train a classification model to recognize different types of robot movements

We'll use the bagpy library to process .bag files and apply basic machine learning techniques for classification.

Although the examples in this tutorial use data from a Boston Dynamics Spot robot (performing movements like moving forward, sideways, and rotating), you can adapt the code for your recordings.

Data management is a very important element in modern robotics systems. As robots become more advanced, they generate massive amounts of information from various sources, such as sensor readings, system logs, and video feeds, and being able to manage this data effectively can mean the difference between a robot that performs well and one that fails to meet expectations.

In this article, we will compare different data management tools for robotics by looking into their key features, strengths, and the types of use cases they are best suited for. Understanding these differences will help robotics engineers and developers choose the right tool for their specific needs.

In this tutorial, we will create a custom ROS 2 Humble package called rosbag2reduct that records incoming ROS 2 topics into MCAP bag files on a Raspberry Pi and automatically uploads those files to a ReductStore instance with metadata labels. We'll walk through setting up ROS 2 Humble on the Pi, interfacing a USB camera using the v4l2_camera driver, deploying a lightweight YOLOv5 (nano) object detection node (using ONNX Runtime) to produce detection metadata, and implementing the rosbag2reduct node to capture data and offload it. We will also cover installing ReductStore on the Pi, configuring replication of labeled data to a central storage on your laptop (using label-based filters via the web console). This end-to-end guide is structured with clear steps, code examples, and configuration snippets to help you build and deploy the system.