Skip to main content

Scientists Discover Way to Send Information into Black Holes Without Using Energy

3 comments

This New AI Lets Robots Learn Once and Never Forget, Even on Different Machines

In today’s fast-moving manufacturing world, robots are everywhere—from assembling products to packaging goods. But there’s a hidden problem that industries quietly struggle with: every time a company upgrades or replaces its robots, it often has to start from scratch. This doesn’t just mean buying new machines—it also means reprogramming them completely.

Even robots designed to do similar tasks can’t easily share instructions. Why? Because each robot has a different physical structure. Some have longer arms, others have different joint angles or movement limits. As a result, a task programmed for one robot usually doesn’t work on another. This leads to wasted time, higher costs, and slower adoption of new technologies.

But now, a major breakthrough from researchers at EPFL’s Learning Algorithms and Systems Laboratory (LASA) could change everything.

A Smarter Way to Teach Robots

The research team has developed a new robotic control framework called kinematic intelligence. This innovative approach allows robots to learn a task once and then share that knowledge across different machines—no matter how different their designs are.

Instead of programming robots line by line, the system begins with a simple idea: teach the robot the way humans learn—through demonstration.

A human performs a task, such as placing an object, pushing it, or even throwing it. Using motion-capture technology, the system records these movements in detail. But rather than copying the exact motion, it converts the demonstration into a general movement strategy.

This is the key difference. Instead of telling a robot exactly how to move, it teaches the goal and logic behind the movement.

Adapting to Different Robot Bodies

Once the system has a general strategy, it needs to adapt that strategy to different robots. This is where kinematic intelligence becomes powerful.

Each robot has its own physical limitations—how far its joints can move, how fast it can operate, and which positions it must avoid to stay stable and safe. The researchers created a structured way to classify these limitations for different robots.

The system then automatically adjusts the learned movement strategy so that each robot can perform the same task within its own capabilities. In simple terms, every robot understands the task in its own “language,” but still achieves the same outcome.

This ensures not only flexibility but also safety—something that is critical in real-world environments.

Real-World Testing: One Task, Three Robots

To test their system, the researchers conducted an experiment in an assembly-line setting.

A human demonstrated a multi-step task:

  • Push a wooden block off a conveyor belt

  • Place it onto a workbench

  • Move it to a table

  • Finally, throw it into a basket

This sequence may sound simple, but it involves precise coordination and timing.

Using kinematic intelligence, three completely different commercial robots were able to perform this exact sequence. Despite having different designs and movement abilities, each robot successfully completed its assigned steps.

Even more impressive, the system remained stable when the roles of the robots were changed. Each robot could adapt and handle different parts of the task without needing new programming.

This shows that the system is not just flexible—it is robust and reliable.

Why This Matters

This breakthrough could have a huge impact on industries around the world.

1. Reduced Costs and Time
Companies won’t need to reprogram robots every time they upgrade their hardware. A single demonstration could be reused across multiple machines, saving both time and money.

2. Easier Deployment
Currently, deploying robots requires skilled engineers and programmers. With kinematic intelligence, even non-experts could teach robots by simply demonstrating tasks.

3. Greater Sustainability
Instead of discarding old programming when robots are replaced, skills can be transferred. This reduces waste and makes robotic systems more sustainable.

4. Faster Innovation
As new robots are developed, they can quickly learn existing tasks without starting from zero. This accelerates technological progress.

The Future of Human-Robot Interaction

The researchers are not stopping here. Their next goal is to make robots even more intuitive to use.

Imagine a future where you can simply tell a robot what to do using natural language—like giving instructions to a human assistant. Combined with kinematic intelligence, this could allow robots to understand both what to do and how to do it.

This has exciting applications beyond factories. In homes, robots could assist with daily chores. In healthcare, they could support caregivers. In logistics, they could adapt quickly to changing demands.

Removing the Barrier of Expertise

One of the biggest challenges in robotics today is the need for technical expertise. Programming robots is complex and requires specialized knowledge.

Kinematic intelligence aims to remove this barrier.

The vision is simple:
The user provides the idea, and the robot figures out the execution.

This shift could democratize robotics, making it accessible to small businesses, startups, and even individuals.

A Step Toward Scalable Robotics

As robotic technology continues to evolve, new hardware platforms are constantly being introduced. This rapid evolution creates a problem—skills developed for one generation of robots quickly become outdated.

By enabling seamless skill transfer, kinematic intelligence ensures that knowledge is not lost when hardware changes. This makes robotic systems more scalable and future-ready.

It also aligns with the broader goal of creating adaptable, intelligent machines that can operate in dynamic environments.

Conclusion

The development of kinematic intelligence marks a significant step forward in robotics. By allowing skills to transfer across different robots, it addresses one of the field’s most persistent challenges.

This approach not only improves efficiency and reduces costs but also opens the door to a more intuitive and human-centered way of interacting with machines.

In a world where automation is becoming increasingly important, innovations like this could redefine how we work with robots—not as tools that need constant programming, but as partners that can learn, adapt, and grow with us.

The future of robotics may no longer be about building smarter machines—but about teaching them smarter ways to learn.

Reference: Sthithpragya Gupta, Demonstrate once, execute on many: Kinematic intelligence for cross-robot skill transfer, Science Robotics (2026). DOI: 10.1126/scirobotics.aea1995

Comments

Post a Comment

Popular

Scientists Discover Way to Send Information into Black Holes Without Using Energy

For years, scientists believed that adding even one qubit (a unit of quantum information) to a black hole needed energy. This was based on the idea that a black hole’s entropy must increase with more information, which means it must gain energy. But a new study by Jonah Kudler-Flam and Geoff Penington changes that thinking. They found that quantum information can be teleported into a black hole without adding energy or increasing entropy . This works through a process called black hole decoherence , where “soft” radiation — very low-energy signals — carry information into the black hole. In their method, the qubit enters the black hole while a new pair of entangled particles (like Hawking radiation) is created. This keeps the total information balanced, so there's no violation of the laws of physics. The energy cost only shows up when information is erased from the outside — these are called zerobits . According to Landauer’s principle, erasing information always needs energy. But ...
3 comments
Read more

Black Holes That Never Dies

Black holes are powerful objects in space with gravity so strong that nothing can escape them. In the 1970s, Stephen Hawking showed that black holes can slowly lose energy by giving off tiny particles. This process is called Hawking radiation . Over time, the black hole gets smaller and hotter, and in the end, it disappears completely. But new research by Menezes and his team shows something different. Using a theory called Loop Quantum Gravity (LQG) , they studied black holes with quantum corrections. In their model, the black hole does not vanish completely. Instead, it stops shrinking when it reaches a very small size. This leftover is called a black hole remnant . They also studied something called grey-body factors , which affect how much energy escapes from a black hole. Their findings show that the black hole cools down and stops losing mass once it reaches a minimum mass . This new model removes the idea of a “singularity” at the center of the black hole and gives us a better ...
Post a Comment
Read more

How Planetary Movements Might Explain Sunspot Cycles and Solar Phenomena

Sunspots, dark patches on the Sun's surface, follow a cycle of increasing and decreasing activity every 11 years. For years, scientists have relied on the dynamo model to explain this cycle. According to this model, the Sun's magnetic field is generated by the movement of plasma and the Sun's rotation. However, this model does not fully explain why the sunspot cycle is sometimes unpredictable. Lauri Jetsu, a researcher, has proposed a new approach. Jetsu’s analysis, using a method called the Discrete Chi-square Method (DCM), suggests that planetary movements, especially those of Earth, Jupiter, and Mercury, play a key role in driving the sunspot cycle. His theory focuses on Flux Transfer Events (FTEs), where the magnetic fields of these planets interact with the Sun’s magnetic field. These interactions could create the sunspots and explain other solar phenomena like the Sun’s magnetic polarity reversing every 11 years. The Sun, our closest star, has been a subject of scient...
Post a Comment
Read more