Marian Vittek

Wonder why they are so focused o teaching this robots to fight humans... : ) Modern humanoid robots are advanced mechatronic systems that integrate actuated joints, perception sensors, and AI-driven control architectures to replicate aspects of human mobility, dexterity, and interaction. Current Capabilities (2025): Locomotion: - Bipedal walking with real-time balance correction using force-torque sensors, IMUs, and dynamic gait planning. - Running and jogging at controlled speeds (e.g., Boston Dynamics’ Atlas demonstrates parkour and backflips). - Stair and obstacle traversal via visual SLAM (Simultaneous Localization and Mapping) and LIDAR depth perception. Manipulation & Dexterity: - Multi-DOF (Degrees of Freedom) arms with tendon-driven or series-elastic actuators enabling fine manipulation. - Pinch grasp, power grip, and in-hand object reorientation with tactile feedback from capacitive/force sensors. - Industrial humanoids (e.g., Figure 01, Tesla Optimus) can handle tools, stack boxes, or perform pick-and-place operations. Perception & Sensing: - Multi-modal sensor fusion: stereoscopic cameras, depth cameras, 3D LIDAR, ultrasonic sensors, and haptic skins. - Real-time object recognition, human pose estimation, and semantic scene mapping using onboard AI accelerators. Human-Robot Interaction: - Speech recognition and natural language processing for voice-based commands. - Expressive face displays or actuated facial features for non-verbal communication. - Gesture recognition through vision-based tracking. Autonomy & AI Control: - Whole-body motion planning with model predictive control (MPC) for stability in dynamic environments. - Reinforcement learning-based locomotion and manipulation policies trained in simulation and transferred to hardware. - Task-level autonomy (fetching objects, assisting in logistics, or assembly) with limited reasoning abilities. Power & Hardware: - Lightweight composite frames with high torque density actuators. - Battery packs (typically Li-ion, ~1–2 hours runtime) with swappable modules. - Edge computing hardware (NVIDIA Jetson Orin, custom ASICs) for real-time AI inference. This is the current technical baseline. Near-term development is focused on improving energy efficiency, hand dexterity, robust outdoor mobility, and safe human-robot collaboration. Credits: Evolving AI #AI #Humanoids #Robots #RoboSapiens

506

94 Comments

Learn the differences between motion profiles! 💨 A real demonstration of engineering skills. In each robotic arm, in each legged robot, you will discover actuators that make the robot move! 🦾 Let's compare two motion profiles: rectangular and trapezoid. - A rectangular motion profile is like a car with abrupt starts and stops, ideal for situations where speed is crucial. 🚙 - A trapezoid motion profile involves gradual acceleration and deceleration, offering a smoother ride, perfect for scenarios where precision and smoothness are essential, just like handling a pint of beer with care (MAB Robotics I love your way of showing the difference! 🍻) P.S. Can you name different applications for these two motion profiles in robotics? ~~~ ♻️ Repost to help 1 robot find a new workplace.

2,271

117 Comments

Bringing Motion to Life with 9-DOF IMU + ESP32-C3 Matrix A compact system using a 9-DOF IMU sensor with an ESP32-C3 microcontroller, that tracks and visualizes real-time orientation data on both an inbuilt LED matrix and an OLED display. It captures pitch, roll, yaw, and showcases omni-directional movement with impressive accuracy and responsiveness. Perfect fit for applications in: 🔸 Robotics – for dynamic balance and motion feedback 🔸 Drones – for flight stabilization and real-time telemetry This tiny yet powerful setup is a game changer for anyone working on AIoT and autonomous systems! Reach us to build your custom projects at: io@CodingScientist.com GitHub: github.com/AnbuKumar-maker YouTube Demo: https://lnkd.in/gAghwjg2 #IMU #ESP32C3 #DroneTech #Robotics #AIoT #MotionTracking #OpenSourceHardware #CodingScientist

237

9 Comments

Understanding C++ concepts is essential for developing a software stack with ROS2. Major components of Nav2 also require C++ deeper understanding of C++ before integrating a custom algorithm into the ROS2 ecosystem. #ros2 #robotics #cpp #robots #nodes

109

1 Comment

🤖 𝗥𝗼𝗯𝗼𝘁𝘀 𝘁𝗵𝗮𝘁 𝘄𝗮𝗹𝗸 𝗹𝗶𝗸𝗲 𝗵𝘂𝗺𝗮𝗻𝘀? 𝗪𝗲’𝗿𝗲 𝗻𝗼𝘁 𝗱𝗿𝗲𝗮𝗺𝗶𝗻𝗴 𝘀𝗰𝗶-𝗳𝗶 𝘄𝗲’𝗿𝗲 𝗰𝗹𝗶𝗺𝗯𝗶𝗻𝗴 𝗿𝗲𝗮𝗹𝗶𝘁𝘆. Humanoid just crushed the hardest obstacle in robotics: stairs.⚠️ No LIDAR, no code just vision, decisions, and fast legs. The question isn’t if it works it’s why we still don’t use it. When I asked a robot to bring me a beer from the basement, I wasn’t kidding. Most humanoids are still hanging on lab hooks. But this one? It climbed over a century-old staircase with no code, no map, no railings. It uses only cameras and real-time vision no LIDAR, no GPS, no pre-recorded motion. And guess what? It works outdoors, in sunlight, shadows, and slanted angles. That’s not sci-fi. That’s skill. Boston Dynamics did flips. Tesla showed renders. This? This is actual walking. Unpredictable terrain. Real physics. Real engineering. ✅ Vision-Based Autonomy No 3D maps, just real-time vision and reaction! ✅ Sub-200ms Reaction Time Adapts faster than most humans can blink! ✅ 94% Stair Climb Success RateTested in unpredictable outdoor conditions! ✅ 20kg Payload Capacity Built to carry tools, goods, or even your beer! So why aren’t we using these in factories, warehouses, or delivery zones yet? Because walking is hard. Because showing off is easy. Because real-world ROI is brutal. What do you say? Legs vs Wheels: Is the biped dream a distraction… or a revolution? Let’s talk. I’m ready for a robot that brings beer and value. Are you? 🚨 I post at 8:00, 11:30 and 17:00 (Berlin time) 3x daily, 100% real robotics. No fluff. No filters. No fakes. 👊 If you’re serious about automation: Follow me now. Hit the 🔔. Don’t wait. 📌 PS: LinkedIn hides most posts. 👉 Join my group or miss out: https://lnkd.in/e9skpAF #Robotics #Automation #Industry40 #AI #HumanoidRobotics #RobotInnovation #RealWorldAI Best regards, #Cobotuli ULMO Consulting & Marketing Video credits : Seen at JOBTOROB via 'Instagram' https://lnkd.in/eiRP9nCg

220

38 Comments

A couple of months ago, Aleksander Sadowski posted a demo of a 3D gear generator web app created with FreeCAD. Here is an analogy written in pure Python in under 200 lines of code. It is based on CadQuery and Streamlit. Want to know how to build similar applications? Read the post of Maksym Valyn on SplineCloud: https://lnkd.in/eRGFd43U #generativedesign #engineering #python #streamlit #cadquery #splinecloud

148

32 Comments

Imagine where we can monitor and analyze vibrations in real time, enabling us to optimize maintenance schedules, detect anomalies, and prevent critical failures. With the combination of Nordic BLE MCU and a high-precision 9-axis Sensor, this vision becomes a reality through a cutting-edge vibration logging data logger. Nordic Semiconductor MCU: The Nordic BLE MCU, renowned for its low power consumption and robust connectivity, serves as the core of this IoT solution. With its advanced Bluetooth Low Energy capabilities, it enables seamless communication and data transfer, even in challenging environments. The MCU acts as the brains behind the data logger, handling sensor data acquisition, processing, and wireless transmission. 9-Axis Sensor: The high-precision 9-Axis Sensor, equipped with an accelerometer, gyroscope, and magnetometer, provides accurate and comprehensive data on vibrations and motion. This sensor's multi-axis capabilities enable us to capture intricate details of movements, rotations, and changes in magnetic fields, providing valuable insights into the dynamics of the monitored equipment. Vibration Logging Data Logger: By combining the power of the Nordic BLE MCU and the 9-Axis Sensor, we can create a compact and portable data logger that collects vibration data in real-time. This logger can be attached to various equipment, machinery, or structures to continuously monitor vibrations and record relevant parameters such as amplitude, frequency, and duration. Real-Time Monitoring and Analysis: The data logger wirelessly transmits the collected vibration data to a central hub, where it can be processed, visualized, and analyzed in real-time. This allows maintenance teams and engineers to gain valuable insights into the health and performance of the monitored equipment. By detecting anomalies or deviations from normal behavior, proactive maintenance can be scheduled, minimizing downtime and optimizing operational efficiency. Predictive Maintenance and Beyond: With the valuable data collected over time, advanced analytics techniques such as machine learning algorithms can be employed to identify patterns, predict potential failures, and optimize maintenance strategies. The synergy between the Nordic BLE MCU and the 9-Axis Sensor opens up endless possibilities in various industries, including manufacturing, aerospace, automotive, and infrastructure. By harnessing the power of IoT, we can revolutionize maintenance practices, improve safety, and unlock unprecedented operational efficiencies. Are you ready to embark on the journey of IoT innovation with Nordic BLE MCU and 9-Axis Sensor? Let's collaborate and reshape the future of maintenance and reliability! #IoT #NordicSemiconductor #BLEMCU #9AxisSensor #VibrationMonitoring #DataLogger #PredictiveMaintenance #Innovation #Technology #DigitalTransformation

32

1 Comment