Eduard R.

🌟 Exciting Update from IFRA-Cranfield! 🌟 Last August, Dr Seemal Asif CENG FHEA MIET and our researcher Mikel Bueno Viso had the opportunity to present our latest research at the ROS-Industrial Developers Meeting (ROS-Industrial Europe). The presentation covered our groundbreaking work on ROS 2-Framework for a Flexible and Reconfigurable Robot Cell Design. 🔍 About the Research: Today's robot manipulators are often designed for specific, prolonged tasks, which limits their flexibility and reconfiguration capabilities. Our research at IFRA-Cranfield aims to overcome this limitation by developing a ROS 2-based software framework that supports hardware/software-agnostic and fully reconfigurable robotic systems. This framework facilitates: - Seamless integration of various robots, automation tools, and AI. - Consistent and intuitive control of robot arms, irrespective of their manufacturer or model. - Fast reconfiguration of robotic systems to adapt to changing production demands and market dynamics. 🎯 Our goal in presenting this work was to share the latest advancements in ROS 2 for the simulation and control of robot manipulators. You can explore more about our work through the IFRA-Cranfield/ros2_SimRealRobotControl repository on our GitHub page, which includes ROS 2 packages for the simulation and control of various robot arms. 📽️ Watch the full presentation on the ROS-Industrial YouTube channel: https://lnkd.in/eQMni3pJ It’s a pleasure for us at IFRA Cranfield to contribute to the ROS 2 community with our research, and we look forward to continued collaboration with ROS-Industrial. 🔧 Call to Action: We encourage the Robotics & Automation Research Community to join us in advancing intelligent and flexible robot simulation and control technologies using ROS 2! #Robotics #Automation #ROS2 #IFRACranfield #Manufacturing #Innovation #RobotControl #ReconfigurableSystems #Research Cranfield University

16

Utilizando la máquina virtual proporcionada por WSL con Debian, para compilar y simular el dron adecuado para trabajar correctamente con algoritmos escritos en Lua. Making use of the virtual enviroment provided in WSL with the Debian distro, in order to compile and simulate a fitting drone to work correctly with Lua scripts. #lua #ardupilot #debian #wsl #drones #development 🪁

14

🔓 Welcome to read the paper "Analyses of Key Variables to Industrialize a Multi-Camera System to Guide #Robotic #Arms" in Robotics MDPI, by Pablo Puerto, Ibai Leizea, Imanol H. and Asier Barrios from BRTA 👉Read more at: https://lnkd.in/e7MXFn3s MDPI

3

I have decided to simulate a modified mockup variant of the mig-29, but due to financial and time constraints the inlet shockwave did not transverse throughout the entire domain. As you can see from the gif the parts of the mig-29 that explicitly result in shockwaves are geometries that although mimic the trajectory of a particle that are thrown in either a curved or straight line but then experience a disturbance of trajectory as expected of oblique shockwaves. The other cause is the taper angle of the mig-29's actual main wings, they need to be narrow and sharp in order to have a descent Coefficient of drag during supersonic flight. If you want to have absurd fluid questions answered, sim-consultants.co.uk is welcome for you to use. Entirely OpenFOAM and Alibaba Cloud based we will answer concerns. #CFD

1

Simon Watson and I are recruiting a #phd student on "Design for Robot Verification". Please share with anyone who is interested in applying. This research will investigate how formal verification tools and techniques can be embedded into the RoboAlchemy design toolchain to help accelerate robotic adoption into industry. Using NASA’s Formal Requirements Elicitation Tool (FRET) to express requirements, we will explore the robustness of software verification tools in the context of modular and reconfigurable hardware systems. Specifically, we will analyse how the composition of robotic architectures may impact the verification of the desired components. https://lnkd.in/eBCi4xUv

48

1 Comment

Check out Active-HDL’s Post Simulation Debug mode in this tool tip video. We’ll show you how to access the mode, how to set up #simulation settings, and how to get post-simulation data. We will also show you which simulation and #debugging tools are available while in the Post Simulation Debug mode. Access the video here: https://lnkd.in/e5aqfy-c   Also, why not try the Post-Simulation Debug mode and other Active-HDL features for yourself. You can request a FREE evaluation license for a fully functional evaluation version of the tool here: https://lnkd.in/e5x7s83b   Follow us and be sure not to miss out on notifications about webinars, product launch news and tool tips. #SystemC #SystemVerilog #Verilog #VHDL #designverification #EDA #FPGA #fpgadesign #fpgasimulation #fpgaverification #hdlsimulation #rtldesign #ActiveHDL

6

It's great to see constant progress 👏 Autonomous mobile robots are becoming more common encoder applications. The RB-Theron from Robotnik Automation uses the 15T/H encoder, a high-performance feedback solution in a low-profile package to ensure better precision. Check the product https://lnkd.in/ggiNTXK #encoder #robots #autonomousvehicles

10

1 Comment

The Medium article from Jose Crespo, linked below, highlights serious vulnerabilities in Tesla’s Supercharger network. Despite expanding to 50,000 stations worldwide, the network relies on a shortest-path model that may struggle with unexpected events like road closures or traffic jams. While effective under ideal conditions, this model can cause issues when drivers deviate from planned routes, such as when Motorways are closed following an accident. While the Trip Planner makes real-time adjustments, it doesn’t address the network’s ability to adapt to unexpected route changes. A 'reachability graph' model, along with k-domination algorithms, could offer a more resilient solution, ensuring better station placement and more options for drivers. https://lnkd.in/eUFTpj9w

20

3 Comments

I am delighted to share our latest paper entitled "Assessment of Panorama Photogrammetry as a Tool for Long-Range Deformation Monitoring." I wish to extend my deepest gratitude to my esteemed colleagues, Luis García-Asenjo, Raquel Luján García-Muñoz, and José Luis Lerma, for their invaluable contributions to this research. This study investigates Panorama Photogrammetry (PPh) as a potential method to collect 3D information in a vast area for deformation monitoring over long distances. Particularly, this study focuses on areas with measuring restrictions, i.e., inaccessible objects and distances above 0.6 km. PPh is investigated in a well-controlled and high-accuracy geodetic reference frame in Cortes de Pallas, Valencia, Spain. I invite you to read the full article through the link provided below: https://lnkd.in/eF7pGMXQ

100

2 Comments

How can RL + MPC be used together to improve controller performance and generalization? There is a significant amount of interest in the robotics community to blend in the strengths of models based methods with reinforcement learning. Here are some ways it is currently done: * RL reference + MPC controller In this case RL is essentially a navigation policy that is conditioned on environment observation. A well trained RL policy can significantly outperform a heuristic driven planner in this case. The advantage of utilizing MPC as the controller is maintaining safety and constraint satisfaction. * MPC reference + RL controller Many RL policies struggle to converge due to sample inefficiency and sparse rewards. In some current research, MPC is used as a dynamically consistent reference generator that the RL policy learns to track and robustify. The problem with this approach is that it requires calling the MPC optimization in the training loop, which increases the training time significantly. The alternative would be to leverage a library of offline trajectory optimization data. * RL high level policy + MPC with parametric uncertainty RL policy can be used to determine the value of an uncertain parameter in the lower level MPC problem. In this case both controllers are trained together, and RL explores how different values of the uncertain MPC parameter affects the closed loop control performance. * RL for warm starting MPC Many MPC problems can be hard to compute in real time as the number of constraints increases. An RL policy trained on many MPC solutions conditioned on observations can produce quality initial guesses to warm start the MPC effectively. #mpc #rl #robotics #reinforcementlearning #roboticsresearch

278

5 Comments

Simulink with super-powers...

5

Month ago I've written the post about why I prefer applied tasks. Let's look at this paper https://lnkd.in/eUtrjQh4. It has 2 key points: 1. We find that training on only 20 digit numbers with a single GPU for one day, we can reach state-of-the-art performance, achieving up to 99% accuracy on 100 digit addition problems.  2. "we do not actually explore any natural language tasks" Am I understand it correctly? -11 scientists -4 universities -2 independent labs -64 exaFLOP of computations 2+months of work ...all of that only for 100-digit summator????

3

Gaussian Elimination ⭐ Solving 3 x 3 System of Equations ☀️ Linear Algebra https://lnkd.in/eh3fmUHK 👉 Fore more, see the playlist Mathematics for Engineers: https://lnkd.in/ekeDH6aC

8

2 Comments

Attention robotics enthusiasts! Check out this blog by my colleague Abhishek Shankar on Learn how you can use custom ROS messages to publish your detection results over ROS with this blog post! https://lnkd.in/gajqsvTK #ROS #Robotics #ObjectDetection 

9

Some great insights from this morning's VDMA ROBOLUNCH webinar on Generative AI & its impact on robotics & automation 🤖 Key speakers Rainer Bischoff, Arne Nordmann, & Markus Wuensch highlighted how AI is set to transform the robotics industry! Arne, from NEURA Robotics made a particularly intriguing point: while generative AI excels at tasks we enjoy like art, music, and writing, it could also tackle tasks we dislike, such as housework 🗑️ For example, if you ask ChatGPT to take out the trash or make a drink, it replies, "I'm unable to physically take out your garbage, but I can certainly provide some tips to make those tasks easier." This underscores the need for AI to have a physical form & appropriate sensors to perceive & interact with its environment effectively. AI & Robotics, when combined, create a powerful synergy with the potential to revolutionize various industries! For more on this topic, below is an article highlighting the advantages & applications of these ground-breaking technologies ⬇️ #robotics #automation #innovation #AI Additae Digital

5

In this tutorial, we'll dive into decision-making (selection) control statements in C++ and explore their various types with example programs. From if, if-else, nested if, to switch statements, you'll learn how to use these control structures effectively to make decisions in your C++ programs. Whether you're a beginner or looking to sharpen your skills, this video has you covered. Here, we have covered the following programs: 1. Program to check if an integer is positive or negative using if-else statement. 2. Program to print the day of the week using switch-case statement. https://lnkd.in/g8DSTt55 Like, share, Subscribe. #coding #beginners #programming #btech #diploma #BCA

2

🌟 A Day in the Life of a MathWorks EDG Intern 🌟 Ever wondered what it's like to intern with the Engineering Development Group at MathWorks? Here's a sneak peek into a typical day: 🕣 8:30 a.m. - Pancake breakfast with fellow interns! 🥞 🕘 9:00 a.m. - Diving into code and solving overnight build issues 🕤 9:30 a.m. - Daily team catch-up to discuss projects and achievements 🔟 10:00 a.m. - Working on a feature to enable communication between a quantum computer and MATLAB 🕛 Noon - Quick lunch break headed to the park with the running club 🏃 🕐 1:00 p.m. - Back to coding and getting valuable feedback from mentors 🕝 2:30 p.m. - Fun social event with other interns, like a river punting treasure hunt 🕟 4:30 p.m. - Code review feedback and getting the "Ship it" signal! 🕠 5:30 p.m. - Wrapping up the day with burgers at a nearby pub 🍔 Have any questions? Drop them below! #MathWorksInterns #Engineering #SoftwareEngineer

1