Ravi Rathnam

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

Read how AGSOTEC demonstrates cloud-based continuous integration with virtual testing and simulation in Simulink. #Automotive #CloudIntegration

3

In the Hartree-Fock method, orthogonalization of the basis turns out to be the most demanding process, since it involves operations with very large matrices. X = UZ−1/2U† where X, U and Z are matrices of size NxN. Fortunately, #GPUs are very good for these types of tasks. In the image below I show the results for the Beryllium atom using FuNQC (Fully Numerical Quantum Chemistry) the code I developed for my PhD. The left side image shows the execution time using #SYCL to compute the orthogonalization of the basis using the GPU, compared with the CPU implementation (right side). The execution times may not be very different, but when atoms with more electrons are involved that could make a big difference.

2

I made some very small tweaks for Carla 0.9.15 ROS Bridge and ROS 2 Humble compatibility. Unlike other versions this one has an updated docker build file: https://lnkd.in/eY_b55M2 #docker #ros2 #CarlaSimulation

10

1 Kommentar

🌟 Exciting Update! 🌟 🚀 Just completed the LangChain for LLM Application Development course from DeepLearning.AI! Gained hands-on experience with LangChain, learning to build and optimize applications using large language models. Key Takeaways: Models: Understanding different model architectures and their applications. Prompts: Crafting effective prompts for optimal model performance. Parsers: Implementing parsers to process and analyze outputs. Memory: Utilizing memory to enhance the context and coherence of responses. Chains: Creating chains to streamline complex workflows. Agents: Deploying agents for automated decision-making processes. Evaluation: Techniques for evaluating and improving model performance. Excited to apply these skills in real-world projects! #AI #MachineLearning #LangChain #LLM #DeepLearning #AIApplications

4

Industry 4.0 touted by the European Union 🇪🇺 since the early 2010 is officially "over", since Smart Automated Factories in Germany cannot compete in this New Industrialization era, where Robotics AI and Machine Intelligence dominates. I call this Nexus as Industry 5.0 of Human-Robot Collaborative and Human-centric Autonomy powered by Robotics AI with Machine Intelligence, making Human workers more productive as CollabAI Teaming. The Chinese understands Robotics and Automation, while the Americans mastered Artificial Intelligence - This is where Singapore aces in bringing the best Global Talents here to do Disruptive Technology as Innovation 💡 Do join my Team to combine this Mastery SkillsFuture of both Robotics and Artificial Intelligence as Machine Intelligence, to advance our Manufacturing and Supply-Chain Operations, alongside many other sectors. Industry 5.0 starts with my Leadership Initiative, please support ... even I am just another very ordinary Singapore-born 55-years old Singaporean Scientist Engineer in Singapore 🇸🇬

44

23 Kommentare

Yesterday's post showcased Altair Inspire's "Conformal Pattern" tools in the Implicit Modeling tab. Today, it only seems fitting to share a video on the "Conformal Lattice" capability. The first step is to choose a lattice type: (1) surface-based like a gyroid or other TPMS; (2) strut-based using unit cells like body-centred cubic, octet; or (3) planar lattices such as hexagon, diamond, triangle, etc.. You then proceed by selecting a parametric (NURBS) surface to conform the lattice to. This transforms the lattice such that it follows the curved surface, avoiding awkward cuts to lattice unit cells. https://lnkd.in/eFkybkGw #onlyforward #CAD #Design #conformallattice #tpms #implicitmodeling #additivemanufacturing #3dprinting

25

1 Kommentar

I’m responsible for trajectory control, the last mile before autonomy. Before we get there, reliable localization, a robust navigation stack, and seamless component integration are essential. The video demonstrate the following features:- 1. Row Following: The robot tracks the planned path (GREEN LINE) while the trajectory controller (BLUE LINE) maintains smooth path alignment. 2. Row Switching: Row switching becomes more complex as rows narrow. We purposely designed the track this way to challenge our nav stack. In the video, you’ll see the robot performing single or multiple 3-point turns to re-align with a new row - requiring precise local optimum odometry and control. Current Setup: Environment: Outdoor and uneven terrain Application: Agricultural and offroad nav. Robot Form Factor: Ackermann steering—although a differential drive setup would be simpler. Contributors:- Mohammad Fairus Khalid : Team lead. Bukhary Ikhwan Ismail : Hardware & sensor selection, localization, nav, trajectory control and tuning. Mahir Sehmi : Sensor assembly and calibration, localization, mapping, path planning, nav, and tuning. Hishamadie Ahmad: Simulation and robot modeling. Ts. Shahrol Hisham Baharom: Robot setup, electronics & electrical.

51

12 Kommentare

Model-based design and automatic code generation has been a paradigm shift in #embedded software engineering. It enables hardware-agnostic development, early validation through simulation and much more. However, oversimplification via graphical modeling has also led to an explosion of unmaintainable code that hurts projects and drives users back to text-based coding. What if there was a way to balance the benefits of model-based design while maintaining control over code quality and working seamlessly with modern collaborative development workflows? Learn more about a code-centric model-based tools in this webcast, and why they should be in your arsenal. https://lnkd.in/eGfWfhrb #embeddedcomputingdesign #etas #ascet #mbd #mcu #ecusoftware

16

What is sensitivity analysis and how does it help your design? Sensitivity analysis enables you to identify key parameters that impact the design. So if you’re in early phase of product development and wondering how to enhance your design, check out the project released by MathWorks Simscape Team. The project has models and workflows like sensitivity analysis that improve robot mechanical design. https://lnkd.in/gGe2ekh6 #sensitivityanalysis #multibodysimulation #mechanicaldesign #robotdesign #articulatedrobot #automation

9

Bridging the gaps between mechanical and electrical design is crucial for accelerating innovation in electronic systems. In Part 2 of our blog series, we explore how integrated ECAD-MCAD collaboration tools like NX and Xpedition are transforming the design process. Breaking down silos between design disciplines is key to staying competitive in today's fast-paced electronics industry. Read the full blog post to learn more about the power of integrated ECAD-MCAD collaboration and share your own experiences in the comments. [https://sie.ag/4DyYEZ]

6

In the rapidly evolving field of automated driving, ensuring functional safety is of paramount importance. ISO 26262 provides a structured approach to achieve this, and MATLAB offers robust tools to streamline the process. Here's a concise workflow for implementing ISO 26262 in your automated driving projects using MATLAB. By following these steps and best practices, you can effectively manage the complexities of developing safe and reliable automated driving systems. MATLAB provides the necessary tools to support each phase of the ISO 26262 workflow, ensuring your project meets the highest safety standards. 🔗 If you are looking to enhance your skills on this topic, check out this Applying Model-Based Design for ISO 26262 - MATLAB & Simulink training (https://lnkd.in/eCHurPkC) offered by MathWorks 🔗 For a more detailed guide, check out this technical article (https://lnkd.in/eBxyFFPT). #AutomatedDriving #IS026262 #MATLAB #FunctionalSafety #Simulink #AutomotiveEngineering #SafetyStandards #ADAS #ModelBasedDesign

54

Battery manufacturing is hard. But MARCO MELISCHKO 🤖 , Roland Spiess 🤖 , Roland Echter 🤖 , and Marie Kühnast 🤖 do an excellent job in the latest edition of E-Mobility Engineering to explain the process. These are real experts sharing insights in process automation to create battery modules and packs that power electric vehicles. #batterymanufacturing #emobility #automation

11