Home Blog

ELECFREAKS Micro:bit XGO Rider Self-Balancing Robot

0

micro:bit is a wonderful platform for building and programming educational robots. The ELECFREAKS Micro:bit XGO Rider is a self-balancing robot powered by it. It is ready for graphical and Python programming. This robot is compact enough to fit on the palm of your hand. It can be controlled with a smartphone or micro:bit powered joystick.

The XGO Rider ships assembled. It has brushless motors and all-metal servos for smooth performance. The battery lasts up to 3 hours.

[where to get it]

Protoclone Synthetic Human Robot with Artificial Muscles

0

This is Protoclone from Clone: it is a musculoskeletal android, which happens to be one of the most anatomically accurate robot in the world so far. It uses 1000 myofibers for movement. Its makers are sharing regular updates all the time. This video shows this the Protoclone robot in 4K:

Wheeled Bipedal Robot Locomotion Controller

0

Wheeled bipedal robots are nothing new. These robots can move faster on even ground and are in general more agile than regular bipedal robots. This video shows a locomotion controller for wheeled bipedal robots that enables them to rotate while going up and down. They can also stay stable on an angled surface.

Wheeled Bipedal Robot

As you can see in this video, the robot can jump up a 20cm step.

[HT] [credit: C.-S. Hao]

Hiwonder xArm AI Vision Robot Arm for Python/Scratch Coding

0

Learning Python and coding in general doesn’t have to be boring for kids. Robots can make it a lot more fun. Take the Hiwonder xArm AI Vision Robot Arm for instance: it is a Python/Scratch programmable robotic arm with WonderCam and WonderEcho modules for color/tag/facial recognition and voice control. The xArm AI has smart servo bus servos for smooth motion.

THEMIS V2 1.6m 40DoF Humanoid Robot Debuts

0

This is the THEMIS V2: a 40DoF humanoid robot with 200 TOPs of computing power, 6DoF arms and 7 DoF end effectors to handle and manipulate objects. These robots used to be available in 3 models, though we are not sure if that still applies to this new one. The below video shows THEMIS V2 in action.

Modular Robot Controller for Contact Rich Manipulation

0

Robots are capable of holding heavy items. Moving them securely is not always straightforward, depending on the object at handle. This video shows a modular robot controller that allows 4 robots applying 32 Newton force to hold and carry a heavy stone. More robots can be added to handle even more complex tasks.

Controller for Contact-rich Manipulation

[HT]

HOVER: Neural Whole-Body Controller for Humanoid Robots

0

This is HOVER: a neural whole-body controller for humanoid robots developed by NVIDIA, Carnegie Mellon University, UC Berkeley, UT Austin, and UC San Diego. It is based on goal-conditioned reinforcement learning for humanoid control. The main idea is to unify different control modes into a single policy, which existing methods handle separately. HOVER unifies kinematic position tracking, joint angle tracking, and root tracking via a multi-mode command space with masking mechanisms.

This approach outperforms task-specific policies in simulation across 12 metrics. This was tested on a Unitree H1 robot. You can read about it at the link below.

[HT]

Falcon Mini Smart Bi-Copter with 4K Video, Reverse Flight Option

0

This is the Falcon Mini: a clever v-shaped bi-copter with app connectivity, 4K camera, and 34-minute flight time. It has tilt-rotor technology and twin servo motors to adjust the angles of the props and maintain stability. Because it has only two propellers, it doesn’t make as much noise.

AIPER Surfer S2 Solar Robotic Pool Skimmer

0

Here is another robot that helps keep your pool clean. The AIPER Surfer S2 is an app connected pool skimmer with smart obstacle avoidance that can work in various swimming pool types. It lasts up to 35 hours on battery for continuous operation. With SolarSeeker tech, this robot ensures charging through the day.

Autonomous Human-Robot Interaction with a Model Trained on Operator Data

0

Here is a robot that can imitate its operator for easy control. Most of the time, teleoperated robots are designed to perform expressive interactions with humans. In this case, it is done autonomously with a model trained on operator data. As the researchers explain:

Our approach learns to predict continuous operator commands through a diffusion process and discrete commands through a classifier, all unified within a single transformer architecture.

This video shows how it works.

Autonomous Human-Robot Interaction via Operator Imitation

[HT]

More Robots: