Meet the robot bird that leaps into the air. Researchers at EPFL have developed RAVEN, a robot inspired by birds’ ability to seamlessly navigate land and air. With multifunctional legs, RAVEN can jump into flight, walk, and hop over obstacles—just like its avian counterparts. Jumping take-offs are not only faster but also more energy-efficient than traditional methods, making RAVEN ideal for tackling complex terrains. Could this bird-inspired design pave the way for more versatile multi-modal robots? #Robotics #drones #Innovation #BioInspiredDesign
关于我们
Your Source for Cutting-Edge Coverage in Mobile Robotics and Autonomy. Circuit is dedicated to delivering comprehensive and up-to-date coverage of the latest developments in the dynamic world of mobile robotics and autonomy. With a team of experienced journalists and industry experts, we strive to provide in-depth analysis, breaking news, insightful interviews, and engaging feature articles that keep our readers at the forefront of this rapidly evolving field.
- 网站
-
https://www.circuit.press/
Circuit的外部链接
- 所属行业
- 互联网出版
- 规模
- 2-10 人
- 总部
- New York
- 类型
- 私人持股
地点
-
主要
US,New York
Circuit员工
动态
-
The first TRUE end-to-end drone surveying solution? Drone surveying has long promised efficiency and accuracy, but ground control setup has remained a persistent challenge—until now. Wingtra has launched WingtraGROUND, an all-in-one solution designed to simplify the process, integrating field setup, aerial mapping, and data processing into a single workflow. For surveyors, the challenge isn’t just capturing aerial data—it’s ensuring that data is accurate. Traditionally, this has meant juggling multiple tools and manually syncing information, leaving room for errors. WingtraGROUND addresses this by guiding users through setup step by step, automatically syncing data, and cutting setup time by a factor of four. "Until now, even with the best drone, you still needed expertise in various third-party tools to ensure accuracy," says André Becker, Senior Product Manager at Wingtra. "By bringing everything into one ecosystem, we’re removing a major barrier to reliable drone data collection." The shift could have far-reaching implications, making high-accuracy drone mapping more accessible to industries that rely on precise geospatial data. #DroneSurveying #AerialMapping #GIS
-
Flapping-wing drones have been around for years... but the X-Fly Bionic Bird takes a different approach. Designed to mimic the motion of a real bird, it combines mechanical wing flapping with onboard stabilization for controlled flight. A six-axis gyro and G-sensors help counteract the vibrations caused by flapping, keeping it steady even in windy conditions. It has a 150-meter range, a replaceable battery for extended use, and a flexible frame designed to withstand crashes. Whether it’s a novelty or a step forward for bio-inspired flight remains to be seen!
-
The moto-robot that drives, jumps, and balances—all on its own, thanks to AI. Developed by the Robotics and AI Institute, this agile robot drives, turns, jumps, and even pulls off precision tricks—all powered by reinforcement learning. From smooth landings to track-stands, every move is autonomously controlled. Does this robot have a commercial future? #robotics #innovation
-
Meet the Wheeled Legged Drone! Researchers at City University of Hong Kong have unveiled a hybrid aerial-ground robot that moves seamlessly between flight and wheeled mobility. Equipped with a reconfigurable single-wheeled leg, it maintains stability in stance mode and achieves high-speed cruising with much higher endurance than a typical quadcopter. The robot has promising applications in surveying and search and rescue.?Where else could it be used? #robotics #drones
-
?? Robots that push with precision Researchers from ETH Zürich have developed a learning-based controller that enables a quadrupedal robot with an arm to push and reorient unknown objects with high accuracy. Using constrained reinforcement learning, the system adapts to different object properties—mass, material, size, and shape—while achieving a 91.35% success rate in simulation and over 80% on hardware. The robot dynamically adjusts its pushing strategy, ensuring stable and contact-rich interactions without prior object knowledge.
-
How do drones handle extreme electromagnetic fields??? The Syddansk Universitet - University of Southern Denmark’s Digital and High-Frequency Electronics team put specially designed drones to the test in a high-voltage lab, assessing their resilience and performance under intense electrical conditions. With support from Fachhochschule Kiel, these experiments push the boundaries of drone reliability for power line inspections and other critical infrastructure applications. #drones #innovation
-
Bridging the Gap Between Simulation and Reality in Humanoid Robotics Achieving agile, whole-body motions in humanoid robots remains a challenge due to the mismatch between simulation and real-world dynamics. Traditional methods often involve labor-intensive tuning or result in overly cautious movements. Enter ASAP (Aligning Simulation and Real Physics), developed at Carnegie Mellon University—a two-stage framework that enables highly agile and coordinated humanoid motions by leveraging real-world data to fine-tune policies trained in simulation. Stage 1: Pre-train motion tracking policies in simulation using human motion data Stage 2: Deploy in the real world, collect data, and train a delta action model to compensate for dynamics mismatch Tested across multiple transfer scenarios, ASAP significantly improves agility and reduces tracking errors, paving the way for more expressive and dynamic humanoids. #HumanoidRobotics #AI #SimToReal #MachineLearning #Robotics
-
This robodog aims to remove humans from hazardous work environments. The Robot Teleoperativo 2 project out of the Dynamic Legged Systems Lab has developed an advanced teleoperation system designed to operate in high-risk environments, minimizing dangers to human workers. By integrating cutting-edge technologies in tele-locomotion, tele-manipulation, and remote human-robot interaction, the system enables precise, dual-arm control for complex tasks in hazardous conditions. #robotics #innovation
-
What Dung Beetles Can Teach Us About Robotics... Dung beetles are expert multitaskers, coordinating all six legs to roll dung balls across varying terrains. But how do they adapt their motor commands to handle different conditions? Researchers from Cornell University have studied this behavior to develop a neural-based control system for robots. By integrating a central pattern generator (CPG), pattern formation network (PFN), and robot orientation control (ROC), they created a robotic dung beetle—ALPHA—capable of walking and rolling balls of different weights and textures on uneven ground. The findings offer insights into sensory-motor coordination, with potential applications for multitasking robots navigating complex environments. #Robotics #Biomechanics #AI #Automation