Enterprise

AgiBot robots can now learn skills on the factory floor

Robotics company deploys real-world reinforcement learning system

Published

on

Robotics company AgiBot has successfully deployed its Real-World Reinforcement Learning (RW-RL) system on a pilot production line with Longcheer Technology.

The company specializes in embodied intelligence, and the project marks the first application of RW-RL in real industrial robotics.

It connects advanced AI innovation with large-scale production, signaling a new phase in the evolution of intelligent automation for precision manufacturing.

Precision manufacturing lines have long relied on rigid automation systems that demand complex fixture design, extensive tuning, and costly reconfiguration.

Where AgiBot’s RW-RL system comes in is addressing such pain points. It enables robots to learn and adapt directly on the factory floor.

Within just minutes, robots can acquire new skills, achieve stable deployment, and maintain long-term performance without degradation.

The system also autonomously compensates for common variations, such as part position and tolerance shifts

Meanwhile, during line changes or model transitions, only minimal adjustments and standardized deployment steps are required. This dramatically improves flexibility while cutting time and cost.

Moreover, AgiBot’s system allows for flexible reconfiguration. Task or product changes can be accommodated through fast retraining; such solution exhibits generality across workplace layouts and production lines.

This milestone signifies a deep integration between perception-decision intelligence and motion control. And it represents a critical step forward unifying algorithmic intelligence and physical execution.

Unlike many laboratory demonstrations, AgiBot’s system also achieved validation under near-production conditions. It has completed a full loop from cutting-edge research to industrial-grade verification.

Moving forward, AgiBot and Longcheer plan to extend real-world reinforcement learning to a broader range of scenarios. These include consumer electronics and automotive components, with focus on developing modular and rapidly deployable robot solutions.

Enterprise

New US-China ban might affect 75% of phones, laptops

Companies can no longer use Chinese labs to test their products.

Published

on

The United States is continuing its crusade against Chinese technology today. However, the target now isn’t a company from China but a method important to a lot of non-Chinese brands.

Today, via Reuters, the Federal Communications Commission (or FCC) has unanimously voted to prohibit companies from using Chinese labs to test their electronic devices if they are to be sold for use in the United States. Naturally, this includes smartphones and computers.

Notably, the prohibition doesn’t directly target Chinese brands. However, it will still affect a huge swath of the industry. The FCC estimates that around 75 percent of the entire market are devices tested in labs based in China.

This means that companies who wish to sell future products in the country must move their testing to labs in the United States or other countries that it deems secure. At its current iteration, the prohibition will not affect devices that already earned their certification prior. However, it might prevent them from getting recertified once their current one expires.

Now, the prohibition isn’t an absolute lock just yet. The FCC will allow the industry to submit comments about the proposal. But, with a unanimous vote from the FCC, companies might have to start looking for alternative testing sites if they want to stay operation in the United States.

SEE ALSO: TikTok finally gets a buyer in the United States

Continue Reading

Enterprise

OnePlus has reportedly merged with realme

Both brands were previously rumored for restructuring early this year.

Published

on

OnePlus 13

OnePlus has a problem. For a while now, rumors have swirled about the company’s dissolution. For their part, the company has continued to deny the reports, citing business as usual. Likely to their dismay, the reports just keep coming. Today, sources have hinted that OnePlus has merged with realme.

Back in January, it was rumored that OnePlus would be closing up shop this year. Since the company very quickly denied the rumors, the report hardly made waves. However, a suspected merger with realme is more difficult to debunk.

For one, realme is itself in a very interesting position. Also back in January, realme was reportedly moving back into being a sub-brand of OPPO. Coupled together with the OnePlus debacle, all this internal restructuring seems par for the course.

According to Digital Chat Station on Weibo, OnePlus and realme have already concluded the merger. The two brands have reportedly united their Chinese and international operations under one roof. Likewise, their marketing will be the same. Pete Lau will still be the main head for this new division.

As with anything of this nature, take this with a grain of salt. OPPO, OnePlus, and realme have not issued any official statements concerning a merger or a shutdown for any brand.

SEE ALSO: realme is reportedly going back to being an OPPO sub-brand

Continue Reading

Enterprise

AGIBOT is turning robots into companions for our everyday routines

The era of robots performing cool tricks is over!

Published

on

The era of robots performing “cool tricks” is over.

At its 2026 Partner Conference, AGIBOT moved embodied AI out of the lab and into the real world.

y using a “One Robotic Body, Three Intelligences” architecture, the company launched five new robot platforms and eight AI models to make physical AI a normal part of how we live and work.

Engineering for human environments

AGIBOT believes that for a robot to be a good partner, it first needs a body you can actually rely on.

Take the AGIBOT A3, for example. This 173 cm tall humanoid weighs 55 kg, about the same as a teenager. It uses a magnesium and titanium build to stay strong yet light.

It moves smoothly for 10 hours straight, and if the battery runs low, you can swap it out in just 10 seconds to keep the momentum of your day going.

In the workplace, the AGIBOT G2 Air acts as a single-arm helper that works right alongside people. It navigates narrow doorways and tight office spaces with ease.

This robot actually learns while it works; it records its environment and actions in real-time to help its AI get smarter every single day.

Then there is the D2 Max, the world’s first Level 3 autonomous four-legged robot. It isn’t a toy you control with a remote; it is a partner that explores tough terrain and handles security patrols entirely on its own.

Finally, the OmniHand 3 series brings a human-like touch to these machines. The flagship Ultra-T model mimics almost any hand movement, while the OmniPicker 3 and OmniHand 3 Lite handle the heavy-duty, high-impact jobs that require extra muscle.

8 models driving autonomy

The “brain” of these machines is a closed-loop system that helps them move, think, and talk.

To master movement, the Behavioral Foundation Model (BFM) allows a robot to copy human actions just by watching a short video.

Another model, the GCFM, lets the robot react to your voice or actions in real-time, which makes its movements feel natural instead of stiff.

To tackle complex tasks, AGIBOT uses a massive dataset called AGIBOT WORLD 2026, a library of real-life situations from homes and factories.

This library helps robots plan out long lists of chores without getting confused. They even use a “digital twin” system called Genie Sim 3.0, where robots practice new skills in a virtual world before trying them in the real one.

On top of that, the WITA Omni model helps the robot understand your feelings, allowing it to talk and move like it’s having a true conversation

Scalable deployment

The robots are becoming a part of our daily lives. By using the MEgo system to collect data easily, AGIBOT is making it simpler for these machines to learn how to help us in shops, warehouses, and our own homes.

As these robots start showing up in our lives, the technology feels less like a complicated machine and more like a companion that helps us grow.

Continue Reading

Trending