Thursday, January 10, 2019

UBTECH's Walker Robot

Walker is one of newest robots from UBTECH Robotics. Below is just a few of the features and technologies used in its development.

1.Flexible walking on complex terrain: With gait planning and control, Walker can achieve stable walking on different surfaces including carpet, floor, marble, and more. Walker can also adapt to complex environments such as obstacles, slopes, steps, and uneven ground.

2.Self-balancing: When Walker is disturbed by external impact or inertia, it can automatically adjust its center of gravity to maintain balance.

3.Hand-eye coordination: Walker’s hands offer seven degrees of freedom to flexibly manipulate objects. By combining its hands with its own perception, Walker can also position dynamic external objects while adapting to uncertain conditions in real-time.

4.U-SLAM navigation and obstacle avoidance: UBTECH Simultaneous Localization and Mapping (U-SLAM) uses environmental information to avoid obstacles and determine Walker’s best path through a dynamic environment.

5.Face and object recognition: Walker has powerful machine vision capabilities to detect and recognize corresponding faces and objects in complex background environments.

6.Smart home control: Walker can help users control common household equipment such as lighting, electrical appliances and electrical sockets, enhancing safety, convenience, and comfort.

With so much innovative technology packed into its humanoid robot body, Walker has the intelligence and capabilities to make a helpful impact in any home or business in the very near future.

Founded in 2012, UBTECH is a global leading AI and humanoid robotic company. In 2018, UBTECH achieved a valuation of USD$5 billion following the single largest funding round ever for an artificial intelligence company, underscoring the company’s technological leadership.

Wednesday, January 9, 2019

Finally, a Do-It-All Robot Arm That’s Actually Affordable

If you want a versatile robot arm, today’s market really only offers two options: expensive industrial robots, or glorified toys. Low-end models may look similar to “real” robot arms, but they don’t usually have the accuracy or repeatability to do actual work. The new Hexbot, however, is designed to give you the best of both worlds.

Hexbot just launched on Kickstarter, but has already reached more than three times the $50,000 funding goal. It’s easy to see why; Hexbot is a small, but capable, modular robot arm that costs just $299 through the Kickstarter Special. That price puts it near the bottom of the market, but it has the kinds of features and specs you’d normally only find on mid-level robot arms.

Machine learning leads mathematicians to unsolvable problem

A team of researchers has stumbled on a question that is mathematically unanswerable because it is linked to logical paradoxes discovered by Austrian mathematician Kurt Gödel in the 1930s that can’t be solved using standard mathematics.

The mathematicians, who were working on a machine-learning problem, show that the question of ‘learnability’ — whether an algorithm can extract a pattern from limited data — is linked to a paradox known as the continuum hypothesis. Gödel showed that the statement cannot be proved either true or false using standard mathematical language. The latest result appeared on 7 January in Nature Machine Intelligence1.

“For us, it was a surprise,” says Amir Yehudayoff at the Technion–Israel Institute of Technology in Haifa, who is a co-author on the paper. He says that although there are a number of technical maths questions that are known to be similarly ‘undecidable’, he did not expect this phenomenon to show up in a relatively simple problem in machine learning.

John Tucker, a computer scientist at Swansea University, UK, says that the paper is “a heavyweight result on the limits of our knowledge”, with foundational implications for both mathematics and machine learning.

Monday, January 7, 2019

Machine Learning for Kids

This tool introduces machine learning by providing hands-on experiences for training machine learning systems and building things with them.
It provides an easy-to-use guided environment for training machine learning models for classifying text, numbers or recognising images.
This builds on existing efforts to introduce and teach coding to children, by adding these models to Scratch (a widely used educational coding platform), allowing children to create projects and build games with the machine learning models that they've trained.