AI Video Tech Tracks Fast-Tracks Robot Training

AI Video Tech Tracks Fast-Tracks Robot Training

NEWNow you can listen to News articles!

One of the greatest obstacles in the development of humanoid robots is the large amount of training data required. Teaching machines to act as humans require massive video sets. Collecting this data is expensive, which requires a lot of time and is difficult to climb. This challenge has slowed progress to make robots useful in everyday environments, such as homes, hospitals and offices.

Register for my free Cyberguy report
Get my best technological tips, urgent security alerts and exclusive offers delivered directly to your inbox. In addition, you will get instant access to my definitive scam survival guide, free when it joins me Cyberguy.com Information sheet.

China launches the center to train more than 100 humanoid robots simultaneously

A slide with information about how Vidar works

Vidar says that their training methods use videos to program robots in 20 minutes. (Vidar)

Vidar de Shengshu transforms humanoid robot training

Shengshu Technology has introduced Vidar, abbreviation for video dissemination for action reasoning. Instead of trusting only in the infinite hours of the physical world, Vidar generates synthetic training environments from a small amount of real video. By combining real data with video generated by AI, Vidar makes training more efficient, scalable and affordable.

A slide with information about how Vidar works

Vidar USA video to train robots to perform real world tasks. (Vidar)

How to Vidar Use Video AI to accelerate robots training

Vidar works by decoupling the perception of control. First, use Shengshu’s Vidu video model to learn from real and synthetic videos. Then, an agnostic task system called Anypos translates that knowledge into motor commands for robots. This modular configuration allows faster training and easier implementation in different types of robots.

Unlike traditional methods that require robots to physically interact with the world to learn, Vidar can simulate complex and realistic scenarios virtually. Surprisingly, it only needs about 20 minutes of training data, between 1/80 and 1/1200 of what the main models require. This efficiency allows climbing robot training to levels never seen before.

The Chinese technology firm shares robot training secrets with the world

A slide with information about how Vidar works in real world applications

The reproduction and implementation of the real world of Vidar with the video model. (Vidar)

Real world applications in humanoid robots

Vidar is more than a simple research tool. Its design means that robots can quickly adapt to new tasks and environments. That could unlock real world applications in elderly care, home assistance, medical care and intelligent manufacturing. When closing the gap between simulation and reality, Vidar is positioning humanoid robots as practical assistants instead of futuristic concepts.

The humanoid robot performs medical procedures through remote control

A slide that shows Vidar tests

Anypos-Atara results with video reproduction to perform several manipulation tasks. (Vidar)

What this means for you

For consumers, Vidar about the idea of ​​domestic robot assistants or workplace to reality. Instead of waiting for decades for robots to mature, scalable training could accelerate the implementation in everyday environments. This could mean that robots help it with tasks, support for the care of the elderly or even helping in medical environments earlier than expected.

Take my questionnaire: How safe is your online safety?

Do you think your devices and data are really protected? Take this fast questionnaire to see where your digital habits are. From passwords to Wi -FI configurations, you will get a personalized breakdown of what you are doing well and what you need improvement. Take my questionnaire here: Cyberguy.com.

Click here to get the News application

Kurt’s Key Takeways

Vidar is a milestone in the race towards practical humanoid robots. By combining limited real data with generative video, Shengshu has created a more intelligent and fast way to train physical. The approach addresses the cost, efficiency and scalability at the same time, three factors that have long delayed robotics.

Would you appreciate a humanoid robot in your home if you could help with daily tasks, or do you still feel too futuristic? Get us knowing in Cyberguy.com.

Register for my free Cyberguy report
Get my best technological tips, urgent security alerts and exclusive offers delivered directly to your inbox. In addition, you will get instant access to my definitive scam survival guide, free when it joins me Cyberguy.com Information sheet.

Copyright 2025 Cyberguy.com. All rights reserved.

Kurt “Cyberguy” Knutsson is a award -winning technological journalist who has a deep love for technology, equipment and devices that improve life with their contributions for News & News Business Startzing Mornings in “News & Friends”. Do you have a technological question? Get the free Kurt’s free newsletter, share your voice, an idea of ​​the story or comment on Cyberguy.com.

Leave a Reply

Your email address will not be published. Required fields are marked *