Luca Marchionni

After reading Andrej Karpathy’s post on X, I’m increasingly convinced that AGI will only be complete when it includes some form of physical systems. Integrating physical components is key to unlocking the full potential of AGI. Check out his thoughts: https://lnkd.in/eRDXsnZT

8

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

Take a look at this demo video. A robot equipped with 360-degree servo wheels and controlled by an Arduino can serve as a versatile platform, particularly suited for research in zero-gravity environments and for educational purposes. Unlike traditional robots, which rely on fixed-direction movement, robots with 360-degree servo wheels offer omnidirectional motion, allowing them to maneuver smoothly in any direction—forward, backward, sideways, or even diagonally—without needing to rotate their body. In zero-gravity environments, such as on space stations or in low-gravity celestial bodies, precision and flexibility in movement are critical. A robot built with 360-degree servo wheels can navigate such environments efficiently by adjusting its orientation and speed on the fly. This capability makes it an excellent candidate for research rovers designed to carry out tasks like inspecting equipment, collecting data, or performing experiments where humans may not have direct access. The Arduino microcontroller acts as the core system of the robot, managing inputs from sensors and controlling the servos. The servo motors provide continuous rotation, which can be precisely adjusted using pulse width modulation (PWM) signals generated by the Arduino. With this control system, the robot can execute complex movement patterns—essential in microgravity research where fine navigation is paramount. Additionally, this type of robot can be used for educational purposes, providing students with hands-on experience in robotics, programming, and systems design. By working with sensors like gyroscopes and accelerometers, students can learn how to stabilize and control the robot in environments where traditional gravitational forces are absent. The platform can also simulate space exploration missions, offering an interactive way to study how robots can perform tasks on distant planets or asteroids. Building such a robot requires components like high-torque 360-degree servo motors, an Arduino board (such as an Arduino Mega or Uno), a motor driver for servo control, and a chassis that can house the electronics. Additional sensors like ultrasonic, infrared, or camera modules can be integrated to provide autonomous functionality, enabling the robot to react to its environment. Ultimately, this robot represents an adaptable research tool, ideal for use in space exploration and education, where traditional methods of locomotion are impractical. Its omnidirectional movement gives it a unique advantage in environments with minimal gravitational pull, allowing it to perform complex maneuvers with ease. #ArduinoRobot #OmnidirectionalRobot #360ServoWheels #ZeroGravityResearch #SpaceExploration #EducationalRobotics #ResearchRover #ArduinoProject #STEMEducation #RoboticsInSpace #ServoMotorControl #MicrogravityTech #RoboticsResearch #ArduinoEngineering https://lnkd.in/gXwebWWs

3

Describing Function and Event Based Analysis of a General Class of Piecewise Feedback Control Systems The analysis of nonlinear systems can be greatly improved using computer software. In this work the Sridhar nonlinearity is described, and a new parameterization is given that allows to systematically obtain a large number of the classical piecewise nonlinearities as particular cases. The calculation of limit cycles using the describing function (DF) approach for the case of autonomous nonlinear control systems and an event-based simulation method are presented and compared. Finally, a user-friendly, graphically oriented interactive tool based on Sysquake is presented. In this way the students can easily assimilate some concepts of introductory courses in nonlinear control such as the behavior of piecewise linear systems, stable and unstable limit cycles, event-based simulation and the DF method. https://lnkd.in/dUeNFrjZ

14

💥💥💥 Full-Order Sampling-Based MPC for Torque-Level Locomotion Control via Diffusion-Style Annealing Abstract Due to high dimensionality and non-convexity, real-time optimal control using full-order dynamics models for legged robots is challenging. Therefore, Nonlinear Model Predictive Control (NMPC) approaches are often limited to reduced-order models. Sampling-based MPC has shown poten- tial in nonconvex even discontinuous problems, but often yields suboptimal solutions with high variance, which limits its appli- cations in high-dimensional locomotion. This work introduces DIAL-MPC (Diffusion-Inspired Annealing for Legged MPC), a sampling-based MPC framework with a novel diffusion-style annealing process. Such an annealing process is supported by our theoretical landscape analysis of Model Predictive Path Integral Control (MPPI) and the connection between MPPI and single-step diffusion. Algorithmically, DIAL-MPC iteratively refines solutions online and achieves both global coverage and local convergence. In quadrupedal torque-level control tasks, DIAL-MPC reduces the tracking error of standard MPPI by 13.4 times and outperforms reinforcement learning (RL) policies by 40% in challenging jumping tasks without any training. In particular, DIAL-MPC enables precise real- world quadrupedal jumping with payload. To the best of our knowledge, DIAL-MPC is the first training-free method that optimizes over full-order quadruped dynamics in real-time. 👉 https://lnkd.in/dSDTBzms #machinelearning

1

Substitution ⭕ vs. Complementarity 📶 ✅   Will a #robot replace us? 😰 Such question we often hear at trade shows and during our demo sessions where people from outside get invited.   When I read chapter 12 «Man and Machine» from Peter Thiel's «Zero to One», I realized it’ s a parallel what people have once thoughts about substitution of humans by computers and software.   ‘’Futurists can seem like they hope the answer is yes. Luddites are so worried about being replaced that they would rather we stop building new technology altogether. Neither side question the premise that better computers will necessarily replace human workers. But the premise is wrong: computers are complements for humans, not substitutes.”   I assume, that many people see robots as species that are very similar to humans in their functions and needs. But that is a purely psychological perception.   ‘The situations are very different: people compete for jobs and for resources; computers compete for neither.” Neither do robots. Machines do not have demands. All value goes for people. That means that people can own a robot and have more time to devote to their lives, families, or more work if they would like to.   I personally can imagine a hairdresser getting a loan and buying a haircutting robot, like this https://lnkd.in/er3qpxNC by Shane Wighton. A beauty therapist buying a lash extension robot, from LUUM a nail specialist buying a number of manicure robots and installing them in local malls.   Paradigm of a #beautician switches to the one of an #entrepreneur. Such machines may give a leverage for specialists to get into businesses, that they already know how to operate. They do know the quality and standards of operations. And robots just may help to make operations better in consistency, quality and quantity.   Here is Jane, the #manicurist. Let’s imagine Jane buys a 10Beauty manicure robot or Clockwork automated nail machine and installs them in her salon. Now she may take 2+n times more clients with nearly same payload. Well, her income increases. Let’s imagine that all manicure therapists on the same streets followed Jane’s steps. So the price went down. It attracts more people to go for the services. And the market either grows or becomes the same. It means that Jane will have the same income as before or higher.   Here is Helen. She is a #massage therapist. Her schedule is full. Her most efficient treatment is lymph drainage massage. Now she can make only 2-3 such treatments a day. Other than that is not quite physically possible for Helen, although her hands are quite trained. So what Helen has to offer to her patients may be roboSculptor. Then the therapist can afford to deliver 2 treatments a day by herself, and up to 12 per day may be treated by the robot. That means that Helen will have enough money to pay leasing of the robot and get additional income. Even if people will lose their jobs because of robots, it’s people who will keep the revenue.

5

Our latest work, RuleFuser, demonstrates that intelligently combining learning-based and rule-based planners surpasses the safety performance of either planner individually. Moreover, this fusion approach exhibits improved robustness to distribution shifts. See the post below for more details. You can find the full paper on arxiv: https://lnkd.in/gbytkh_5

52

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 comentario

We are happy to announe that our paper "Leveraging transformers architectures and augmentation for efficient classification of fasteners and natural language searches", in collaboration with Marco Murgia Nino Cauli and Giuseppe Scarpi (R2M Solution), has been published in Discover Computing (the old Information Retrieval journal) by Springer. This is the abstract A primary concern in the realm of mechanical engineering is to ensure the efficient and effective data entry of hardware devices. Fasteners are mechanical tools that rigidly connect or affix two surfaces or objects together. They are small and often different fasteners might look similar; it is therefore a long and prone-to-risk procedure to manually analyze them to classify and store their related information. With the widespread diffusion of AI frameworks in several domains, equipment manufacturers started to rely on AI technologies for these heavy tasks. Automatically classifying fasteners by type and extracting metadata from natural language questions are important tasks that fastener manufacturers and suppliers encounter. In this paper, we address these challenges. To address the first task, we introduce an augmentation methodology that starts with a small set of 3D models representing each of the 21 types of fasteners we aim to classify. This methodology efficiently generates multiple 2D images from these models. Next, we train a vision transformer using the collected data to address a single-label multi-class classification task. For the second task, we introduce a prompt-engineering technique designed for conversational agents. This technique leverages in-context knowledge to extract (metadata field, value) pairs from natural language questions. Subsequently, we tackle a question-answering task to the description fields of the extracted fasteners. Our evaluation demonstrates the effectiveness of both approaches, surpassing the baselines we tested.

31

Meshing in #CFD simulation – is a complex and challenging process! Traditional mesh-based methods in #CFD are computationally expensive and time-consuming too. Our QA-PINN based methods offer a promising mesh-free alternative by leveraging quantum computing. This focused on reducing the number of trainable parameters in the PINN architecture while maintaining accuracy.  This research can advance engineering simulations in many ways such as: Reduced computational cost: QA-PINN can potentially handle complex CFD problems more efficiently than traditional methods. Improved accuracy: QA-PINN can achieve higher accuracy in simulation results due to its enhanced ability to capture complex physical phenomena. Scalability: It is well-suited for large-scale simulations, enabling the modeling of complex systems. This research has the potential to revolutionize #CFD simulations, enabling more efficient and accurate modeling for real-world applications. Abhishek Chopra will be presenting this paper on Quantum-Assisted PINN (QA-PINN) at TODAY at IEEE Quantum Week. #IEEE #Quantum #QML #pinn

32

If we give Robots "Working Memory" systems, how large should their WM buffers be, and how long should entities remain in WM before decaying away? It depends on context. Rafael Sousa Silva explores in his Cogsci 2024 paper: https://lnkd.in/gx4fquHJ In this work Rafael examines robot WM systems that are not bound to human-like restrictions, where every task-relevant entity can have its own cache of relevant features. But just because robots have vast memories doesn't mean these systems don't need limits! In this work, Rafael carefully examines how the upper and lower bounds on both decay and interference parameters are sensitive to (1) contextual factors, and (2) tolerance for risk. Specifically, Rafael analyzes the error modes of WM systems -- what can go wrong if decay and interference parameters are too low or too high? For example, if decay is too large, it becomes increasingly probably that WM will contain "stale" properties that are no longer accurate. If we assume properties go stale following an exponential distribution, we can precisely analyze the relationship between decay rates and our risk tolerance for property staleness! Rafael performs this type of analysis for the upper and lower bounds of interference and decay parameters, ultimately allowing him to formulate "Metarules" for simultaneously reasoning about these upper and lower bounds. This was a surprisingly fun paper -- and one that REALLY benefitted from this year's CogSci reviews, which encouraged Rafa to really flesh out some of this math for the paper's final version. To learn more, read the paper! https://lnkd.in/gx4fquHJ And big thanks to National Science Foundation (NSF) for funding it!

29

Is ROS a useful skill when searching for a job in robotics? Are robotics companies asking for that skill? I analized 1.904 jobs on LinkedIn to answer that question, and the results... will surprise you! ;-) Check the answers in this podcast episode of the ROS Developers Podcast: https://lnkd.in/dPwMW-3b

65

4 comentarios

Companies won't go from 100 developers down to 0 because of AI. They will, however, solve 10x more problems with the same team. 10x more software, 10x higher quality. We may see a net gain in software jobs, not a decrease. Gugu Robotics emerges as the changemaker from the future, creating solutions for the Military, Industries, Medics, Agriculture, Mining, Underwater, Space, and everyday people to conquer uncharted terrains and extreme mundane tasks. As our planet stands upon a precipice of enormous change and transformation, let's embrace innovation and shape a brighter tomorrow together. Ready to elevate your Business? Contact Gugu Robotics to discuss custom builds and experience the future of excellence. --- For more information, Our website is Currently under Repair. Book a Meeting https://lnkd.in/df4vpzHR and Discuss what Solution we build for you. Let's build the future together. Emmy Pencil©

3

This insightful document defines quantum machine learning as the integration of a small-scale quantum circuit with a classical optimizer (e.g., SGD, Adam). As such, understanding the fundamental principles and design of quantum computing is a prerequisite. #machinelearning #quantumcomputing #optimizer

7

Exploring MQTT broker 🔥 on FlowFuse with my homies: 🚀 Revolution Pi 🔥 Raspberry Pi 🖥️ reTerminal from Seeed Studio 🖥️ Raspberry Pi based HMi from EDATEC Video coming soon 🚀- Stay tuned! Follow me ✅ for Tech updates on Automation and IIoT #iiot #mqtt #flowfuse #revolutionpi #raspberrypi #automation #codeandcompile #industry40 #nodered #industrialautomation

124

7 comentarios

Announcement #Sony #Robotics Solution for #AMR / #AGV with #ROS2 !!! If you are interested, please feel free to reach out to us: https://lnkd.in/gXZX4Unc AutonMate: https://lnkd.in/gBs-sbRk Autonomous Mobile Robot (AMR) that assists with picking through collaboration with humans, aiming to reduce labor. AMRs, positioned as the next generation of Automatic Guided Vehicles (AGVs), can travel without guides and move autonomously while avoiding people and obstacles. Robotics packages are already running in the market and actual workspace in Japan!!! #Sony has been working to make a lot of contribution to the mainline based on the requirements from business logics, then we can use #ROS as a user to the business applications. We really appreciate #ROS community and #ROS open source, and we will keep it up #ros #ros2 #robot #mobilerobot #opensource #oss #robotics

451

1 comentario

Our MVP BMO-AI is absolutely lightening fast in conversation response using the latest version of Llama. I love how it’s more realistic in the flow, like a conversation you would have with a human friend. A verbal ping pong game! Hope others are enjoying the latest release too! #AI #Robot #llama3 #startup

4

2 comentarios

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