Foxglove

?? The comma2k19 dataset visualized using Foxglove ?? Comma2k19 offers a rich collection of real-world driving data, captured over 33 hours on California’s Highway 280. This dataset is crafted to fuel innovation in developing fused pose estimators and mapping algorithms. The dataset includes synchronized data from road-facing cameras, GPS, thermometers, and 9-axis IMUs, giving an in-depth view of real-world driving conditions. It also provides raw GNSS measurements for creating precise localization algorithms, along with CAN data that reveals vehicle dynamics essential for advanced control systems. Technically, the dataset is organized into 10 chunks, each containing about 200 minutes of driving data, with processed logs in numpy arrays and global camera poses. It supports sensor fusion, enabling tightly coupled INS/GNSS/Vision optimizers for robust pose estimation. Data is available in raw and processed formats, with standardized timestamps and sensor units. Given the above, we added pose projection directly onto images in a dedicated image annotation topic to visualize the car’s movement and spatial orientation within the frame. This addition is invaluable for building more accurate visual odometry, pose estimation, and scene reconstruction algorithms, as it links 3D positional data with 2D visual contexts, enhancing algorithmic reliability in autonomous navigation systems. Thanks for the awesome dataset, comma.ai. ?? Link in the comments to check out the dataset ??

We're excited to be hosting Foxglove & Friends Holiday Happy Hour next week in San Francisco! Join us on Thursday, December 5th for a chance to connect and mingle with the Bay Area robotics community over food & drinks. Whether you're a Foxglove user, customer, or friend, this is our way of saying thank you for supporting us! We have room for 30 more people! Apply with the link below to secure your spot ??

?? Announcing: Folders to organize your layouts ??? With the release of 2.17 you can now create folders to organize both your personal and shared layouts. Keep your team organized and your layouts tidy by clicking the ? menu next to any layout to create or add it to a folder. Download link in the comments to start organizing your layouts now ??

Happy Friday! We're capping off our Proxie reveal week. Wednesday we dropped our product storytelling video. Thursday we dropped the story of Proxie at Maersk as told by Erez A.. Links to both are in the comments. Today we're leaving you with a slideshow of some of the shots of Proxie in our videos and on our website. We hope you enjoy these! Next week, we're going to highlight our Flywheel consulting program and how we're helping enterprises accelerate their robotics journey. We'll let three of our customers tell those stories as we head into Thanksgiving. We expect to have more exciting stuff to share immediately after Thanksgiving. Then mid-December, we'll share a little more of the story behind the storytelling as we showcase some of the partners that have helped us to make this successful. Stay tuned here... lots more to share.

“It was possible to locate all instances of an issue, and summarize its impact on the robots’ performance within minutes using Foxglove”. Rob Zehner, VP of Engineering, Coco Learn how Coco reduced incident resolution time from hours to seconds using Foxglove. ?? Link in the comments ??

?? Getting started with Foxglove is incredibly simple using the Foxglove WebSocket Bridge app on your iOS device. ?? Connect your iOS device’s sensors—like GPS, accelerometer, and camera—to Foxglove through WebSocket for real-time data visualization. Download the app, input your WebSocket URL, and you’ll be able to monitor live feeds and sensor data on customizable panels in Foxglove instantly. Oh and it comes with all the fun #Actuate2024 stickers and more too! Enjoy ?? Link to the full blog in the comments ??

?? Foxglove 2.17 is available now! ?? Highlights: Play files from start to end as if they’re a single recording. Foxglove now supports merged playback of multiple local files. Simply select Open local file(s), choose multiple files from your system, and view the combined data seamlessly. ?? Organize layouts with folders: Stay organized by creating folders for your personal and shared layouts. Keep your team’s layouts tidy by clicking the ? menu next to a layout and selecting 'Add to folder'. ?? Plot & State Transitions using Publish Time: Plot time-series data and state transitions using publish_time, in addition to the default log_time (receive time) and optional header.stamp fields. This adds flexibility to visualize your data based on your needs. ?? Simplify Plot CSV output: Exporting data from the Plot panel is now easier to work with. Right-click the top bar of any plot and select Download plot data as CSV. ?? Easily copy logs: Copy your current logs directly to the clipboard. A small change that makes your daily workflow smoother and more efficient. This release also comes with: ? Improved the readability of the State Transitions panel. ? The Topics sidebar now denotes which fields use enum types. ? Enums from IDL schemas are now used when displaying values in the Raw Message and State Transition panels. ? Added recording key to recording.created and recording.imported webhook event payloads. ?? and so much more. Lets go! ?? Download and changelog links in the comments ??

?? The Intelligent Manipulation (d-lfd) dataset visualized using Foxglove ?? D-ifd, developed by the Intelligent Manipulation Lab (IMANLAB), is a comprehensive resource for tackling the intricate task of autonomous robotic suturing. Focused on the task of needle insertion into deformable tissues, the dataset captures complex bi-manual operations where one robotic arm inserts the needle while the other manipulates the tissue, aligning actual and desired exit points—a challenge that truly tests adaptability due to tissue deformation. The dataset includes synchronized recordings from three pairs of stereo cameras, precise kinematic data from Intuitive's da Vinci Research Kit (DVRK) arms, and extensive joint-state tracking. This data enables the development and refinement of control algorithms that enhance precision, control, and efficiency in robotic surgery systems. Accurate positioning of each joint is key for fine motor control, allowing the robot to make precise real-time adjustments to counter tissue deformation. While the combined joint states offer an overview (tab "JS Together"), individual plots of each joint state (tab "JS Positioned") provide the precision needed for exact control. Calibration data is also meticulously included, specifying the relative poses of the cameras and their alignment with the robot’s base frames, which further enhances data reliability and interpretation. This dataset is made possible by the contributions of Amir Ghalamzan, Kiyanoush Nazari, Hamidreza Hashempour, Fangxun Zhong ?? Link to the dataset in the comments ?? #Robotics #DataVisualization #DataViz #Autonomous

?? Meet Alvie, the Purdue Lunabotics robot used to place in the top 5 at NASA’s annual Lunabotics competition. ???? Purdue Lunabotics is a team of undergraduate students dedicated to creating autonomous lunar mining robots for NASA’s annual Lunabotics competition. Competing against over 50 universities nationwide, teams aim to build the largest regolith berm within a set time frame, earning points for autonomy, reduced robot weight, efficient power usage, and the quantity of regolith moved. In 2024, Purdue advanced through qualifiers and placed 4th overall. This competition, part of NASA’s Artemis Program, challenges students in mechanical, electrical, and software engineering, advancing skills critical to human space exploration. Check out the full story to learn how Foxglove helped Purdue Lunabotics’ place 4th in their journey to autonomy in space. Full story in the comments ??

“Thanks to Foxglove we're no longer spending time fighting with tooling — we're developing revenue-generating product and helping our customers, which is why we exist.” Charles Quinn, Co-founder and CEO, Greenzie Learn how Foxglove helps Greenzie cut incident triage time from days to minutes. ?? Link in the comments ??