Robotics In Science Projects

Build your first robot in simulation! 👾 📌 If you’re self-learning robotics, this is genuinely one of the better repos to save for later. NVIDIA Robotics released a "Getting Started with Isaac Sim" tutorial series covering everything from building your first robot to hardware-in-the-loop deployment. What's inside? → Building Your First Robot Explore the Isaac Sim interface, construct a simple robot model (chassis, wheels, joints), configure physics properties, implement control mechanisms using OmniGraph and ROS 2, integrate sensors (RGB cameras, 2D lidar), and stream sensor data to ROS 2 for real-time visualization in RViz. → Ingesting Robot Assets Import URDF files, prepare simulation environments, add sensors to existing robot models, and access pre-built robots to accelerate development. → Synthetic Data Generation Learn perception models for dynamic robotic tasks, understand synthetic data generation, apply domain randomization with Replicator, generate synthetic datasets, and fine-tune AI perception models with validation. → Software-in-the-Loop (SIL) Build intelligent robots, implement SIL workflows, use OmniGraph for robot control, master Isaac Sim Python scripting, deploy image segmentation with ROS 2 and Isaac ROS, and test with and without simulation. → Hardware-in-the-Loop (HIL) Understand HIL fundamentals, learn NVIDIA Jetson platform, set up the Jetson environment, and deploy Isaac ROS on Jetson hardware. The progression makes sense: start with basics (build a robot), add perception (sensors and data), generate training data (synthetic generation), develop software (SIL), then deploy to hardware (HIL). Each module builds on the previous one. For robotics teams, this is the path to faster iteration. Simulate first, validate in software-in-the-loop, generate synthetic training data at scale, then deploy to hardware with confidence. 🎓 If this helps at least one engineer to become more fluent in the world of robotics, means a lot to me! 🫶🏼 Here's the course (it's free): https://lnkd.in/dRYdkmdi ~~ ♻️ Join the weekly robotics newsletter, and never miss any news → ziegler.substack.com

A research team incorporated a microphone into soft robotic fingertips to detect fabrics through sound, achieving 97% classification accuracy on 20 common fabrics. The system combines internal vision with audio sensing to create multimodal touch perception. The technique is surprisingly simple but effective—the microphone picks up the acoustic signatures as the fingertip interacts with different fabric textures. Combined with visual feedback from inside the finger, it creates a classification system that matches human-level performance. The embedded video on their site is genuinely impressive—robotic ASMR meets practical manipulation. If you're working on tactile sensing or dexterous manipulation, this multimodal approach is worth exploring. Project page: https://lnkd.in/dxEkqbJ3 Paper: https://lnkd.in/dR2RXUd7 Via Weekly Robotics 349: https://lnkd.in/dhCW3WrR #Robotics #Research #TactileSensing #MachineLearning #SoftRobotics

Why integrate all sensors into ONE drone with low computational cost? For years, we’ve built different drones for different sensors—one for LiDAR, another for thermal, another for mapping. But what if one intelligent drone could do it all—efficiently? Here’s what changes when you move to a multi-sensor, low-compute architecture: ✅ Mission Versatility Switch seamlessly between mapping, inspection, agriculture monitoring, and search & rescue—without changing platforms. ⚙️ Operational Efficiency One drone. Multiple payloads. Fewer deployments, reduced logistics, and simplified training. 💰 Cost Optimization Lower hardware duplication, maintenance overhead, and deployment costs. 🧠 Higher Data Value (Sensor Fusion) Combining LiDAR + RGB + Thermal + Multispectral + IMU + GNSS unlocks context-rich insights that single sensors simply can’t provide. ⚡ Low Computational Cost = Real Edge AI With optimized onboard processing (TinyML, edge AI accelerators, efficient pipelines), we: → Reduce latency → Minimize data transmission → Enable real-time decision-making at the edge 🌍 Impact? One drone becomes a multi-domain intelligence system—scalable, adaptive, and future-ready. One drone. All perspectives. Smarter decisions. Lower cost. Greater impact. #Drones #SensorFusion #EdgeAI #AIoT #Robotics #AutonomousSystems #SmartMapping #PrecisionAgriculture #ComputerVision #EmbeddedAI #LowPowerAI #Innovation #AerialIntelligence

Improvised #magnetometer sensing system test: checking the feasibility of using magnetometers on #quadruped #robot. We took a standard UAV magnetometer payload — the SENSYS - Magnetometers & Survey Solutions #MagDrone R3 — and attached it to a DEEP Robotics (https://lnkd.in/g4vaP2Mz) LYNX M20 robot dog. Some early outcomes: - In general, it works, and the concept looks viable. Our larger targets are clearly visible in the data, although the noise level is extreme. - A quadruped robot is an extremely EM-noisy platform. The biggest issue is not just the amplitude of the noise, which we expected, but its constantly changing frequency across a wide range. This makes a simple low-pass filter practically useless. Each robot leg has three powerful motors, each running at different speeds — and even when they are not rotating, they remain engaged and still emit significant noise. - The next step is to move the sensor far enough from the platform to reduce interference below an acceptable level. For UXO and other man-made object detection tasks, noise below 1 nT is acceptable. Based on our noise analysis, we need at least a 125× reduction. Thanks to the inverse-cube relationship with distance, this means increasing the distance from the nearest motor by about 5×. - In our experimental setup, the distance from the R3 sensor to the nearest motor was approximately 20 cm, so a 1 m separation in the final configuration should be sufficient. - Are UAV payloads suitable for robot dogs? Not really. They are too fragile. Payloads for quadruped robots need to be much more rugged — closer to military-grade. We have already broken plenty of experimental mounts, sensor rods, and booms. Stay tuned. There are many scenarios where UAVs and classic UGVs are not suitable, and we see strong potential for sensor-equipped robot dogs. SPH Engineering

Nav2Bot: ROS 2 Autonomous Navigation in Ignition Gazebo ➡ Differential drive robot simulation using ROS 2 Humble ➡ Autonomous navigation using Nav2 stack ➡ LiDAR-based obstacle detection and environment perception ➡ AMCL-based localization for accurate robot positioning ➡ Global and local path planning with real-time execution ➡ Complete TF tree (map → odom → base_link → lidar_link) ➡ RViz visualization for costmaps, paths, and robot pose ➡ Keyboard teleoperation support for manual control ✨ Why this matters: Autonomous navigation is one of the core challenges in robotics, where a robot must perceive its environment, determine its position, and plan a safe path to a goal without human intervention. This project demonstrates a complete ROS 2 Nav2 pipeline that integrates localization, planning, and control into a unified system. By combining LiDAR data, odometry, and costmaps, the robot can intelligently navigate through unknown environments while avoiding obstacles in real time. These principles are widely used in real-world robotics applications such as autonomous vehicles, warehouse automation systems, delivery robots, and service robotics. 📊 Key Highlights: ✔ Full ROS 2 Navigation Stack (Nav2) integration ✔ LiDAR-based perception and obstacle avoidance ✔ AMCL localization for accurate positioning ✔ Global and local path planning ✔ Real-time costmap generation ✔ Gazebo simulation with realistic robot behavior ✔ RViz-based monitoring and debugging 💡 Future Potential: This framework can be extended to: ➡ Multi-robot navigation systems ➡ SLAM + Nav2 integration for unknown environments ➡ AI-based dynamic obstacle detection ➡ Reinforcement learning for path optimization ➡ Real-world deployment on mobile robots 🔗 For students, engineers & robotics enthusiasts: This project provides a complete hands-on implementation of autonomous navigation using ROS 2, making it ideal for understanding how intelligent robots perceive, plan, and act in real environments. 🔁 Repost to support robotics research & engineering education! #ROS2 #Nav2 #Robotics #AutonomousSystems #Gazebo #Mechatronics #EngineeringProjects #Lidar #RViz #Automation #Navigation #AI #STEM #EngineeringEducation #RobotSimulation

A Reliable Outdoor Localization for ROS 2 Robots with GPS + IMU + Odom I have recently received a lot of questions about setting up GPS in ROS 2 projects, here’s a package I have used a lot for my outdoor robotics projects: robot_localization. It’s the backbone of sensor fusion in my outdoor robotics builds. What robot_localization helps you do: ✅ Fuse GPS (NavSatFix), IMU, and odometry into a pose estimate ✅ Handle coordinate transformations and frames (map↔odom) ✅ Use navsat_transform to convert geographic GPS data to local frames ✅ Configure which sensor readings and axes to trust) ✅ Support for ROS 2 distros like Humble, Jazzy, Rolling Raw GPS is noisy and inconsistent, IMUs drift, odometry slips. robot_localization helps you fuse these signals so your robot knows where it is outdoors. 🔗 If you’re trying an outdoor navigation project, this is a tool you want in your stack! Have you used robot_localization with GPS already? What issues (drift, sensor delay, covariance tuning) have you run into? Let’s connect and share robotics tips 🔽 #ROS2 #Robotics

If I could tell my younger self one thing before starting a career in AI and robotics, it would be this. When I joined EarthSense, Inc. six years ago, I thought AI was my job. Now I know deploying vision and AI on robots in the real world isn't one problem. It's actually 6 problems hiding under a trenchcoat. And you have to master all of them to deliver a real solution. 📷 Sensors & 🧠 Compute An AI system can only be as good as its eyes and its brain. Are you using camera, LiDAR, or both? Monocular or depth? How many can your computer support? NUC or Jetson, and which ones? What's your budget? What's your plan for calibration, exposure, dealing with motion blur? 💾 Data Management 100TB of sensor data sounds great. But how will you upload it, and where? How will you track dates and robot serial numbers? How will you query, download, process, and use it? What data formats do you need? What format conversions do you need to run? ✏️ Labeling & Datasets Can't train AI without labeled datasets. How will you label your data? Who will do your labeling? What tool will they use? How will you adapt to domain shift? How will you review labeling quality? 🤖 Neural Networks Time for AI! What architecture will you use? What training methods? How will you scale training to large compute clusters? How will you deploy inference? How will you optimize for edge? ⚙️ Algorithms Raise your hand if you thought AI was the only hard part. How will you project detections onto depth images? How will you fuse camera, LiDAR, IMU together? How will you implement and validate pose transform logic? How will you build and manage a map of the surroundings? What do you do if two sensors disagree with each other? 👁️ Evaluation, Analysis & Visualization Time for pretty visuals. Where is your system working and failing? When something goes wrong, where do you even start with analyzing it? What is all of your telemetry data telling you? What reliability guarantees can you provide to customers? 🏁 Conclusions In the robot prototyping process, you'll see failures at any step of the pipeline. For this reason, you need to be adaptible. Your group needs to master each of these areas of development, and integrate them smoothly. How do you deal with the complexities of deploying AI in the real world? Follow me for a front-row seat to what it really takes to build AI systems that work in the real world. I share lessons from the field, insights from the lab, and behind-the-scenes from our company as we scale. #AI #FieldRobotics #Robotics #Autonomy #ComputerVision

Flexible pressure sensors are everywhere in prototypes. But scaling them to real-world production? That’s where most projects fail. In our work designing pressure mapping systems for robotic end effectors, we faced two challenges at once: → Build a high-fidelity, flexible sensor array. → Make it manufacturable beyond the lab bench. Here’s what it took: → Material System Selection. Why? We couldn’t just pick the softest or thinnest FSR materials. We needed materials that could survive lamination, mechanical cycling, and environmental stress without losing responsiveness. → Matrixing Without Crosstalk. Why? In a grid of distributed sensors, each node needs to be individually addressable without electrical interference bleeding across rows and columns. We engineered trace geometries and insulative layers to keep signals clean — even under flex and inflation. → Layered Durability. Why? Flexibility often sacrifices lifespan. We designed stackups that maintained elasticity while protecting conductive layers from mechanical fatigue and delamination. → Manufacturing Alignment: Why? Prototyping with hand-aligned layers is easy. Scaling requires layers to be aligned mechanically or laser-cut to tight tolerances, without introducing shifts that ruin sensor performance. It’s not enough to build a working prototype anymore. If you want to move from a concept to something scalable, you have to engineer for: → Mechanical reliability → Electrical integrity → Production repeatability

What if building a full robotics stack didn’t take a team of ten PhDs… or 2 years of development... or $5 million dollars? That’s exactly what the new Advantech Edge AI SDK is changing. For years, integrating ROS 2 + LiDAR + cameras + AI models has been a heavy lift: - Custom drivers - Sensor fusion pipelines (gPTP anyone?) - SLAM tuning - Model deployment - Endless debugging across disconnected systems It wasn’t just hard—it was resource prohibitive. Now? You can do it in a no-code / low-code environment. ✔ Pull ROS 2 packages directly from GitHub ✔ Drop in LiDAR drivers and camera inputs ✔ Run SLAM out of the box ✔ Load and train AI models for perception ✔ Deploy everything on industrial-grade edge hardware All in one unified stack. This is the real shift: We’re moving from building infrastructure → to building intelligence. Instead of spending months wiring systems together, you can now: Simulate Train Iterate Deploy …in a fraction of the time. My take: Systems built in 1 day with this stack can outperform legacy systems that took 10 years to develop. That’s not incremental progress—that’s a step change. LiDAR users—take note. Sensor fusion, SLAM, and perception are no longer bottlenecks. They’re becoming commoditized (read FREE) capabilities. The competitive edge is shifting to: 👉 How fast you can deploy 👉 How fast you can iterate 👉 How intelligently you use the data The barrier to entry for advanced robotics and physical AI just collapsed. And this is only the beginning. #Advantech #EdgeAI #ROS2 #Robotics #LiDAR #ComputerVision #AMR #AGV #PhysicalAI #Automation #AI #SLAM #IndustrialAI