Foxglove

🚨 Announcing: support for AV1, VP9, and H.265. Only in Foxglove! 🚨 With the release of 2.19 Foxglove adds support for AV1, VP9, and H.265, extending codec support beyond H.264. We’re particularly excited about supporting all four leading codecs, both natively and directly in the browser. By embracing the strengths of each codec, Foxglove is delivering unparalleled flexibility, enabling you to choose the optimal solution for your specific needs. You can now choose between the codec being used—“h264,” “h265,” "vp9", or “av1”—and define it in the format field of the foxglove.CompressedVideo message schema. Populate the data field with the appropriate units: NAL units in Annex B format for H.264 and H.265, or OBUs for VP9 and AV1. Each message should contain enough packets to decode exactly one video frame. Link to read the full blog in the comments 👇

🚨 Announcing: support for AV1, VP9, and H.265. Only in Foxglove! 🚨 With the release of 2.19 Foxglove adds support for AV1, VP9, and H.265, extending codec support beyond H.264. We’re particularly excited about supporting all four leading codecs, both natively and directly in the browser. By embracing the strengths of each codec, Foxglove is delivering unparalleled flexibility, enabling you to choose the optimal solution for your specific needs. You can now choose between the codec being used—“h264,” “h265,” "vp9", or “av1”—and define it in the format field of the foxglove.CompressedVideo message schema. Populate the data field with the appropriate units: NAL units in Annex B format for H.264 and H.265, or OBUs for VP9 and AV1. Each message should contain enough packets to decode exactly one video frame. Link to read the full blog in the comments 👇

The Foxglove NVIDIA Robotics Isaac Sim extension enables real-time visualization of robotics simulation data directly in Foxglove. In case you missed our previous post on the extension, we released an NVIDIA Isaac Sim extension that enables seamless visualization of simulation data in Foxglove. The extension automatically detects all cameras, IMUs, and articulations in the simulation stage, making the data—along with the complete Transform Tree—available in Foxglove. In this blog, we'll take a deeper look into the extension's code to understand how it works and explore ways it can be expanded upon. Check it out. Link in the comments👇

The Hugging Face 🤗 Le Robot Unitree H1 datasets visualized using Foxglove. 🤖 The Unitree H1 folding clothes and arranging items datasets, captures a robot performing common daily tasks, offering a rich source of data for advancing robotics, computer vision, and AI research. It includes 38 episodes with a total of 19,000 frames recorded at 50 fps, featuring both 19-dimensional state vectors and stereoscopic RGB images at 1280x720 resolution. Each frame also includes a 40-dimensional motor command vector, enabling precise action tracking. Designed for efficiency, the dataset is stored in Parquet format, making large-scale data processing seamless. We converted it to MCAP for seamless visualization using Foxglove. The visualization no only displays the RGB images but also plots the joint states, adds the Unitree H1 URDF to a 3D scene, and incorporates DepthAnythingV2 in the middle colored set of depth images. Visualize the dataset directly in Foxglove for your self. Links to the dataset and DepthAnythingV2 project in the comments 👇

Another incredible video from Waymo showing their quick reaction times. Would a human have reacted this fast? What if they happened to turn their attention to their kids/phone/stereo for 250ms?

UTFR Driverless Achieves Complete Autonomous Navigation What began as an ambitious vision to develop an autonomous formula student racing car has culminated in a major milestone: for the first time, our end-to-end driverless system successfully navigated an autocross track. This achievement is not just a testament to technical capability but also a reflection of our team's dedication to advancing autonomous systems from concept to real-world application. Our development journey progressed along two parallel threads. One focused on iteratively improving our driverless software stack throughout the season, while the other tackled the integration and refinement of the car’s mechanical, electrical, and firmware systems. On the hardware side, we implemented and optimized critical components such as the emergency braking system, autonomous steering, remote emergency stop, vehicle state, and safety checks. Meanwhile, the software stack addressed key challenges in perception, mapping, planning, and control, ensuring the system could perceive the environment, map the track, plan efficient paths, and reliably execute those plans via the vehicle’s actuators. This collaborative effort between software and hardware ultimately enabled the system to meet the demanding requirements of real-world autonomous racing, bridging the gap between simulation and physical performance. Key milestones in our technical progression: • November 2023: First validation of Emergency Braking System • May 2024: Full EBS implementation on vehicle platform • June 2024: Remote emergency stop system validation • July 2024: Comprehensive DV system validation • September 2024: Major improvements to the perception module • November 2024: Full autonomous navigation of Acceleration and AutoX courses Under the technical leadership of Kelvin Cui, Daniel Asadi, and Youssef Elhadad in the past few years, the driverless section has now established a foundation for future innovation. Our focus now shifts to integrating SLAM capabilities and optimizing performance for upcoming Formula Student competitions in Michigan and Europe. Critical to our development process has been our partnership with Foxglove, whose sophisticated visualization and cloud storage platform has illuminated the intricate behaviours of our autonomous systems, enabling rapid iteration and refined optimization. We extend our gratitude to our sponsors and faculty advisors, whose support and guidance have been instrumental in this achievement. This milestone extends beyond technical achievement – it demonstrates how student engineering teams can push the boundaries of innovation while developing crucial real-world engineering expertise. We are excited to continue on our march to unveil UT25 in due time. https://lnkd.in/gPmKfEEN

Karthik Keshavamurthi, Motion Planning Engineer at Scythe Robotics, shares the full development journey behind their advanced “Return to Home” feature for the Scythe M.52 at #Actuate2024. He takes you through every stage, from gathering crucial customer feedback and defining key requirements to designing, testing, and validating the final solution. Stream the entire talk now. Link in the comments. 👇

Michael Laskey, CTO of Electric Sheep, delivered a simply electric talk at #Actuate2024 on the company’s innovative ES-1 model: a foundational world model that processes time-series data to predict both raw features and interpretable outputs, including semantic understanding, bird’s-eye-view mapping, robot positioning, and traversability assessments. Combined with reinforcement learning ES-1 enables autonomous execution of outdoor tasks such as mowing and trimming. Check it out now. Link is in the comments 👇

Learn how Aescape streamlined debugging from days to minutes with Foxglove. 💜 “The impact of Foxglove was immediate. Within a few weeks of adopting the platform, our debugging cycles became significantly shorter.” Scott Butters, Staff Machine Learning Engineer, Aescape Link in the comments 👇

I’m #hiring a Staff Product Designer at Foxglove! Looking for a designer with 6 years of experience who enjoys working on complex data visualization products. Come design tools for the leading robotics companies in the industry! https://lnkd.in/entDsvwU