Foxglove

SLAM (Simultaneous Localization and Mapping) is crucial for autonomous robotic systems, enabling efficient navigation and mapping in unknown environments. While various SLAM frameworks cater to different sensor modalities, they all focus on two primary objectives: refining a robot's trajectory and simultaneously updating its map. Lidar sensors are particularly valued for their precision in high-speed applications like self-driving cars, providing accurate 2D or 3D point cloud data. Additionally, Inertial Measurement Units (IMUs) are often integrated to provide additional motion data, enhancing system robustness. Modern SLAM frameworks leverage lidar data using graph optimization techniques (e.g., LIO-SAM), filter-based methods (e.g., EKF-SLAM, FastSLAM, Point-LIO, FAST-LIO2), and feature-based methods (e.g., LOAM). This beautiful example dataset was collected by Doncey Albin and team out of the Autonomous Robotics and Perception Group (ARPG) at University of Colorado Boulder, using an AgileX Robotics Hunter SE equipped with a 64-beam Ouster OS1 lidar, LORD Microstrain IMU, and CENet, trained on SemanticKITTI and modified into ROS1, used on the CU Boulder campus as a zero-shot method. Links to learn more in the comments. 👇

“The adoption of Foxglove was swift, with the impact being viscerally clear from the first day, and it continues to deliver value two years into Saronic’s operations.” Vibhav A., Co-founder & VP of Software, Saronic Technologies Check out the Foxglove + Saronic success story 💜 Link in the comments 👇

💜

One of the principal tasks for any mobile robot is being able to move in its environment without colliding with walls, people or obstacles in general. That should be obvious... and that’s where Nav2 comes in. Steve Macenski created #Nav2 and then Open Navigation LLC. It is considered one of the most important frameworks in ground robotics. It’s going to be such a pleasure to chat with Steve live at #Actuate2024 on September 18 and hear him talk about different use cases with Nav2. 🗺️ Get your tickets and join me live at Actuate: https://lnkd.in/guVvt3iV

What does it take to train a robot's mind? Nick Obradovic, Senior Director of Software Engineering at Sanctuary AI is going help answer that question live on stage at #Actuate2024 on September 18. The race in humanoid innovation has felt like it’s really heated up the last couple years so I can’t wait to hear Nick’s answers and find out how he and Sanctuary AI have trained their Phoenix Gen 7.🔥 Use code ADRIANSOCIAL for a 25% discount on your tickets and join me live: https://lnkd.in/guVvt3iV

🔦 Student Spotlight: Jiali Sun on modeling and control of PADUAV 🚁 As a master's student at the Beijing Institute of Technology, with a research focus on novel aerial robot design and control, Jiali's daily routine revolves around controller design, validation in simulation, and experimental testing. Analyzing data to validate whether the design works is crucial. Jiali and team's recent work, accepted for #ICRA2024, introduced a novel 5-DOF aerial platform called "PADUAV" designed to overcome the limitations of traditional tiltrotor vehicles. Novel robot designs often come with novel controller designs that require extensive testing to identify and rectify defects. The simplest and most intuitive way to inspect the controller's performance is by observing the curves —assessing the extent of overshoot, convergence or divergence, and oscillation. Read all about Jiali's work and more from the link in the comments. 👇 A special thanks to Jiali Sun for helping put this article together.

On September 18, live on stage at #Actuate2024, Vijay Badrinarayanan, VP of AI at Wayve is going to teach me and a crowd of robotics developers how Wayve’s groundbreaking research is revolutionizing the journey to embodied AI, particularly in self-driving. 🚘 I’m also keen to learn more about PRISM-1 and LINGO-2 directly from Vijay! If you are too use code ADRIANSOCIAL for a 25% discount on your tickets and come and join us: https://lnkd.in/guVvt3iV

🛸 Foxglove 2.11.0 has shipped! 🛸 Highlight: The Diagnostics Summary and Detail panels now support message converters, so these panels can be used when your data source does not use the standard ROS DiagnosticArray message type. This release also comes with: 👓 Webhook IDs are now displayed in the Webhooks page, viewable by organization Admins, making it easier to manage and debug your webhooks. 🕴️ Hovering over a topic in the sidebar now shows the full topic name ⏱️ The Insights page for organization Admins now allows you to adjust the time range for the data displayed 😮 OMG IDL union message discriminators no longer require a specific or default case defined 💾 Improved playback performance and reduced memory consumption when using multiple 3D or Image panels ⏺️ Added a topic query parameter to the GET /v1/recordings endpoint to filter on the presence of a given topic 💜 and much more... Let's go! 🔥 Download and changelog links in the comments 👇

The resume speaks for itself. It’s an absolute honor to share the stage with Brad Porter, Founder & CEO, Cobot - Collaborative Robotics, Inc, on September 18th at #Actuate2024. I’m sure Brad has more than one hot take to share as we chat about his experiences and opinions on scaling robotics development and the future of the industry. 🌶️ Use code ADRIANSOCIAL for a 25% discount on your tickets and join me live in person: https://lnkd.in/guVvt3iV

If you're in 🇪🇸 Spain 🇪🇸 one month from now check out ROSCon ES 2024 Seville. 🚨Just announced: "Saca el máximo partido a Foxglove empleando tus propios datos!" Presentado por José Luis Millán Valbuena. Acelera tu desarrollo en robótica aprovechando todo el potencial de Foxglove, un software de visualización de primera clase, empleando tus propios datos. En este taller, trabajaremos con rosbags compartidos por ti y la comunidad para mostrar casos de uso reales donde aprenderemos cómo Foxglove puede aumentar tu productividad y la de tu equipo. ¡Y además, es gratis!