Kinegami: a paper on tubular origami

🌟 Excited to Share Our Research on Humanoid Robotics! 🌟 Thrilled to introduce our paper, "Fuzzy Dynamic Gait Pattern Generation for Real-Time Push Recovery Control of a Teen-Sized Humanoid Robot," published in IEEE! https://lnkd.in/gwCidvxR 🤖 What’s the big idea? We developed a fuzzy dynamic gait pattern generator that enables a humanoid robot to recover from unexpected pushes in real time. Unlike conventional approaches relying on pre-planned trajectories, our method integrates the Virtual Force Linear Inverted Pendulum Model (VFLIPM) with advanced sensor feedback, ensuring adaptive and robust push recovery during dynamic scenarios. 🔍 Key highlights: Dynamic Adaptation: Combines pressure sensor data and accelerometer feedback with a fuzzy controller to instantly compute recovery gaits. Real-Time Responsiveness: Adjusts the robot’s Center of Mass (CoM) trajectory based on balance status. Proven Performance: Successfully implemented on the teen-sized humanoid robot, David Junior II, for the RoboCup Push Recovery event, showing a recovery robustness 1.3x stronger than standing mode. 🚀 This research showcases a step forward in real-time humanoid stability control, with applications in robotics competitions, real-world human-robot interaction, and beyond. I’d love to hear your thoughts and connect with others interested in humanoid robotics, fuzzy logic control, or dynamic motion planning! 🌐 #Robotics #Humanoid #FuzzyLogic #PushRecovery #Research #IEEE

Looking for a movie to chill out with? How about Jung_E, one of the four movies discussed in @Scirobotics in 2024. Come for the realistic ZIMBO hybrid leg/wheel vehicle (which looks like DRC-HUBO from the Defense Advanced Research Projects Agency (DARPA) robotics challenge) action sequences, stay for the ethical drama! See the science at https://lnkd.in/g3KMHGY9 

Worth a read: New from MIT SPARK Lab, a mechanism for deriving task dependent scene graphs. Take a set of tasks in natural language, and compute the granularity required to interpret the robot’s surroundings to focus upon those elements of the scene relevant to the task. Published in IEEE Robotics and Automation, and available here; “Clio: Real-Time Task-Driven Open-Set 3D Scene Graphs” https://lnkd.in/guuNRdXT I particularly liked this quote from Luca Carlone; “Search and rescue is the motivating application for this work, but Clio can also power domestic robots and robots working on a factory floor alongside humans.”

🚀 Taking Armatronix to New Heights in ROS2! 🚀 I’m thrilled to share recent updates on my ROS2 project, Armatronix! The robot is now equipped with a functional camera, enabling real-time vision, and its skid-steer locomotion is running smoothly for autonomous movement. 🔍 What’s New: - Provided vision to the robot 🤖 - Enabled locomotion for it to explore the environment. - Physical parameters like Inertia, Collision, Torque, and Mass are fine-tuned, providing realistic dynamics. - Arm kinematics are implemented, bringing precision to manipulation operations. 🌐 Up Next: - Integrating MoveIt2 for trajectory planning and manipulation tasks. - Refining full autonomy with advanced ROS2 navigation and control functionalities. - Applying Deep Learning Algorithms to teach and learn from the surroundings. 🛠️ Tools & Technologies: - ROS2 for end-to-end control - RViz2 for visualization - Custom URDF modeling - Skid-steer plugin for dynamic locomotion - Real-time camera plugin for visual feedback Stay tuned as I bring Armatronix closer to full autonomy! 💡 #Robotics #ROS2 #Automation #AI #MoveIt2 #ROS

The paper "Deblur-GS: 3D Gaussian Splatting from Camera Motion Blurred Images" by Wenbo Chen and Ligang Liu presents a method for reconstructing 3D Gaussian points from camera motion blurred images. This method, called Deblur-GS, jointly optimizes camera trajectory estimation and time sampling to create a sharp radiance field from blurred images. The approach models the problem of motion blur as a joint optimization challenge involving camera trajectory estimation and time sampling, achieving superior performance and rendering quality compared to previous methods. #gaussian #motion #vision #blurredimages #trajectoryestimation #robotics #CNN #transformer #LLM #video ▪️Code: https://lnkd.in/eUD6_PUt ▪️Paper: https://lnkd.in/enQNfR8H ▪️Project: https://lnkd.in/e6PZJxJX

Hello Robotics Enthusiasts, Setting up robots in simulation can be a challenging task, especially with ROS2 adding a new layer of complexity. In my latest video, I walk you through the steps to set up a Universal Robot in ROS2 in just 4 mins, covering key areas like: - Building your workspace - Cloning repositories from GitHub - Moving the Universal Robot in Gazebo using MoveIt 2 If you're eager to enhance your robotics skills, check out the video below. Let’s learn and build together! 🎥 Watch here - https://lnkd.in/gTsTwXsp Happy learning, and keep building awesome projects! #ROS2 #SIMULATION

Simulation-Based Forward Kinematics! 🎥 Following a recent post on solving forward kinematics for a SCARA robot using Denavit-Hartenberg parameters, here’s an alternative method using a simulation-based approach. This technique brings the kinematic model to life, offering a real-time visual representation of the robot’s motion and end-effector trajectory, enabling clear visualization of the end-effector's path. The end-effector coordinates obtained through this simulation method match those calculated using the Denavit-Hartenberg parameters, confirming the accuracy of both approaches in modeling the SCARA robot’s motion. This approach provides an engaging, dynamic view of SCARA robot functionality, with the end-effector's path validated against the DH parameter method. #Robotics #SCARA #ForwardKinematics #Simulation #MechanicalEngineering #MATLAB #Automation

Excited to share that my publication, Soft Bifunctional Utility Module for Robot Sensing and Signaling (S-BUN), which was presented at IROS 2024, is now officially published on IEEE Xplore! This work highlights a novel integration of sensing and signaling for enhanced human-robot interaction. Thankful for the incredible collaboration and support from my team! Read the full paper here: https://lnkd.in/e86_SvP2 #Robotics #SoftMaterials #HumanRobotInteraction #IEEE #IROS2024 #Research”

To solve locomotion problems, the mobile roboticist must understand; ✅Mechanism and Kinematics, ✅Dynamics and ✅Control theory. To create robust perceptual systems, the mobile roboticist must leverage the fields of signal analysis and specialized bodies of knowledge such as computer vision to properly employ a multitude of sensor technologies. Localization and navigation demand knowledge of; ✅Computer algorithms, ✅Information theory, ✅Artificial intelligence, and ✅Probability theory. LEVON ROBOTICS

Becoming a Roboticist Week 2 Day 2 & 3 w/ Dr. John Vial :Code Organization and new Robot Simulation Today I learned how to have ROS (Robot Operating System) create a new package to contain a robot and simulation.  We also created a new GitHub repository to keep our coming work all neat and organized. With our new ROS package created we now need to add a robot and simulation to our package/repo.  To do this we leveraged the open-source community and the Xacro (Zac-crow) tool. The ur_robot xacro is a macro that setups all the parameters needed to make a new robot arm. This is included in the Universal Robot ROS package. Using this macro will create a URDF (Unified Robot Description Format) file from the xacro macro that includes all the information about the robot. Now that we have a robot we need a simulation to put it in. We created a gazebo launch file to create our world and add our robot in. In all we have learned to leverage open-source macros and tools to generate our own robot in our new instance of the simulator.