Robots are going to do many things for us. To handle every day objects, they are going to need a dexterous end-effector. The qb SoftClaw allows robots to handle food items, tools, and everything in between. It has a plug & play design and max payload capacity of 5kg. It has interchangeable fingers.
Robots can find it a bit tricky to grab glass and mirrors. These objects are reflective and make it hard for robotic sensors and grippers to handle them effectively. With Pre-Touch Sensor TK-01 by Osaka University startup Thinker, it is possible for robotic arms to grab and handle mirrors, glass and other reflective objects. It achieves real-time measurements of “distance and position” of objects at high resolution.
A cutting-edge sensor lets robotic hands grip hard-to-see items like glass, mirrors, & roses! Instead of conventional image tech or touch, Pre-Touch Sensor TK-01 by Osaka University startup Thinker Inc. uses infrared & AI for detection.https://t.co/3PEK5rONfD#InnovationJapan pic.twitter.com/z9ODGnPnwU
— japan (@japan) February 7, 2025
The above video shows this robot in action. You can find out more about the sensor here.
Here is a framework that allows for expressive motion sequences in humanoid robot, significantly improving their non-verbal communication. This approach relies on LLMs to generate social appropriate gesture motion sequences. Robots will then execute motions via inverse kinematics.
Apple’s AI researchers introduce EMOTION, a framework for expressive humanoid gestures.
⦿ LLMs interpret social context and generate motions from human demos
⦿ Robots execute motions via inverse kinematics
⦿ Gestures are refined through human feedback
Hardware: Fourier GR-1 pic.twitter.com/B5fxBttfpx— The Humanoid Hub (@TheHumanoidHub) February 7, 2025
This framework was tested with Fourier GR-1. EMOTION can incorporate human feedback. The researchers generated and evaluated various expressive gestures.
[HT]
Robots are going to do many things for us in the future, including common household chores. 7X has already shown robots that can autonomously wash dishes. Their latest video shows how a robot can autonomously fold 3 t-shirts. They claim the total cost for this will be about $7000.
Our @7Xrobotics robot autonomously folds three t-shirts one after the other. Total cost about $7000. We’re making a Windows executable and will post all hardware details. This is the same robot that was autonomously washing dishes earlier. This guy loves to do your housework. #AI… pic.twitter.com/pBH5BthHCu
— Usman Roshan (@Deeplearner2) February 6, 2025
According to Usman Roshan, all hardware details will be posted in the near future.
[HT]
Golf carts are getting smarter all the time. We have already seen a few that can follow players on a golf course. The iXi Self Driving Golf Cart takes it to the next level with voice/gesture recognition and an AI camera. It moves with you, carries your clubs, and tracks your swings. It comes with a large touchscreen display. iXi has a lightweight, portable design.
Writing and drawing robots are nothing new. We have covered a bunch of them that can create a work of art and replicate human-like handwriting. This robot from PNDbotic (Adam) can write Chinese Fu. Adam SP stands 1.67m tall, weighing 620kg. It has 41 degrees of freedom.
It is no secret that robots are becoming more agile. They can run and do all kinds of challenging poses. with ASAP, it is now possible to teach humanoid robots agile whole-body skills. Thanks to this approach, it is possible for a humanoid robot to perform moves by LeBron James and Cristiano Ronaldo. ASAP is a two stage framework that involves pre-training motion tracking policies in simulation using retargeted human motion data.
This is the ToddlerBot: a $6k open source humanoid robot for AI research. It has 30 active degrees of freedom, including 7 DoFs per arm and 6 DoFs per leg. It can lift just under 1.5kg while maintaining its balance. These robots can also integrate with OpenAI’s real-time API for speech to text and text to speech conversations.
In the past few years, we have covered plenty of cute educational robots for kids. The XR Hunter-Ackerman is a Jetson Nano ROS2 robot for more advanced students. It comes with AI vision and LiDAR navigation. It is ready for Python programming. This model comes with a 4GB Jetson Nano. It also has a 2DOF 3D camera and a foldable 7-inch screen.
Plenty of robots are inspired by nature, even fleas and mantis shrimp. Take HeTRM for instance: it is a hyperelastic torque-reversal mechanism for powerful and rapid motions using soft rubber-like materials. As you can see in the below video, the soft gripper is capable of grabbing ping pong balls. This approach can be used to build robots that move on rough terrains.
