MAROKO133 Update ai: Spinning-mass robots that roll and swim could soon achieve insect-lik

📌 MAROKO133 Breaking ai: Spinning-mass robots that roll and swim could soon achiev

An orange wheel rolls across concrete and suddenly jumps, as if it decided to defy gravity on its own. Nearby, a robot undulates through water like a fish. Another twists forward inside a narrow pipe.

These machines look different, but they all move using the same physical principle.

At Clemson University, mechanical engineering professor Phanindra Tallapragada and his team are building robots that rely on centripetal force rather than complex joints, legs, or propellers.

The approach uses an unbalanced spinning mass inside each robot to generate motion.

The idea is simple. When an off-center mass spins fast enough, it creates forces that can push, lift, or twist the robot’s body.

The effect is similar to a washing machine vibrating when wet clothes collect on one side of the drum.

Tallapragada sees these robots as physical expressions of mathematics rather than traditional machines packed with motors and software.

“A lot of robotics today is perceived as designing something with motors, microcontrollers, machine learning and AI– and less importance is given to the dynamics and math, at least in the public perception,” he said.

One force, many motions

That same spinning-mass principle drives several robots developed in Tallapragada’s lab. One of them is a remote-controlled wheel known as the Spin Gyro.

It jumps when the internal mass rotates fast enough to lift the wheel off the ground. Unlike spring-based jumpers, it can repeat the motion almost instantly, making it useful for rough or uneven terrain.

The lab has also built a fish-like robot that swims by transferring energy from a spinning mass into its tail. By adjusting the rotation speed, the robot can dive, turn, or move forward.

Compared to many swimming robots, the system is highly energy efficient and could be used for tasks such as monitoring lakes, oceans, or water quality.

Another robot is designed to crawl through tight pipes. Inside, a spinning mass causes small bristles on its body to rapidly compress and release.

Friction with the pipe walls pushes the robot forward, allowing it to move through spaces as narrow as an inch. Potential uses include inspecting ducts, gas lines, or pulling cables through long pipe networks.

Graduate students and postdoctoral researchers are deeply involved in the work. “I want to be a roboticist, and that’s what motivates me every day– to make robots that do something useful in the world,” said Prashanth Chivkula, who helped develop the swimming robot.

From ground to air

The team is now extending the same physics into the air. Tallapragada has launched a new project aimed at building insect-inspired flying robots.

Instead of conventional motors, spinning masses could drive wings at the extremely high frequencies needed for insect-like flight.

The effort is supported by a three-year grant from the U.S. National Science Foundation.

Beyond Earth, Tallapragada believes the concept could be useful for planetary exploration.

A single robot capable of rolling, jumping, swimming, and flying could navigate icy surfaces, leap through vents, and dive beneath frozen crusts on distant moons in search of liquid water.

“I just like working on these new ideas that are very different from what others are doing and seeing them come to physical action,” Tallapragada said.

“Once they come into action, that gives me more motivation to go back to the whiteboard and come with more models.”

🔗 Sumber: interestingengineering.com

📌 MAROKO133 Update ai: Researchers from PSU and Duke introduce “Multi-Agent System

Share My Research is Synced’s column that welcomes scholars to share their own research breakthroughs with over 2M global AI enthusiasts. Beyond technological advances, Share My Research also calls for interesting stories behind the research and exciting research ideas. 

Meet the author
Institutions: Penn State University, Duke University, Google DeepMind, University of Washington, Meta, Nanyang Technological University, and Oregon State University. The co-first authors are Shaokun Zhang of Penn State University and Ming Yin of Duke University.

In recent years, LLM Multi-Agent systems have garnered widespread attention for their collaborative approach to solving complex problems. However, it’s a common scenario for these systems to fail at a task despite a flurry of activity. This leaves developers with a critical question: which agent, at what point, was responsible for the failure? Sifting through vast interaction logs to pinpoint the root cause feels like finding a needle in a haystack—a time-consuming and labor-intensive effort.
 
This is a familiar frustration for developers. In increasingly complex Multi-Agent systems, failures are not only common but also incredibly difficult to diagnose due to the autonomous nature of agent collaboration and long information chains. Without a way to quickly identify the source of a failure, system iteration and optimization grind to a halt.
 
To address this challenge, researchers from Penn State University and Duke University, in collaboration with institutions including Google DeepMind, have introduced the novel research problem of “Automated Failure Attribution.” They have constructed the first benchmark dataset for this task, Who&When, and have developed and evaluated several automated attribution methods. This work not only highlights the complexity of the task but also paves a new path toward enhancing the reliability of LLM Multi-Agent systems.
The paper has been accepted as a Spotlight presentation at the top-tier machine learning conference, ICML 2025, and the code and dataset are now fully open-source.

Paper：https://arxiv.org/pdf/2505.00212
Code：https://github.com/mingyin1/Agents_Failure_Attribution
Dataset：https://huggingface.co/datasets/Kevin355/Who_and_When
 
 
Research Background and Challenges
LLM-driven Multi-Agent systems have demonstrated immense potential across many domains. However, these systems are fragile; errors by a single agent, misunderstandings between agents, or mistakes in information transmission can lead to the failure of the entire task.

Currently, when a system fails, developers are often left with manual and inefficient methods for debugging:
Manual Log Archaeology : Developers must manually review lengthy interaction logs to find the source of the problem.
Reliance on Expertise : The debugging process is highly dependent on the developer’s deep understanding of the system and the task at hand.
 
This “needle in a haystack” approach to debugging is not only inefficient but also severely hinders rapid system iteration and the improvement of system reliability. There is an urgent need for an automated, systematic method to pinpoint the cause of failures, effectively bridging the gap between “evaluation results” and “system improvement.”

Core Contributions
This paper makes several groundbreaking contributions to address the challenges above:
1. Defining a New Problem: The paper is the first to formalize “automated failure attribution” as a specific research task. This task is defined by identifying the

2. failure-responsible agent and the decisive error step that led to the task’s failure.

Constructing the First Benchmark Dataset: Who&When : This dataset includes a wide range of failure logs collected from 127 LLM Multi-Agent systems, which were either algorithmically generated or hand-crafted by experts to ensure realism and diversity. Each failure log is accompanied by fine-grained human annotations for:
Who: The agent responsible for the failure.
When: The specific interaction step where the decisive error occurred.
Why: A natural language explanation of the cause of the failure.

3. Exploring Initial “Automated Attribution” Methods : Using the Who&When dataset, the paper designs and assesses three distinct methods for automated failure attribution:
All-at-Once: This method provides the LLM with the user query and the complete failure log, asking it to identify the responsible agent and the decisive error step in a single pass. While cost-effective, it may struggle to pinpoint precise errors in long contexts.
Step-by-Step: This approach mimics manual debugging by having the LLM review the interaction log sequentially, making a judgment at each step until the error is found. It is more precise at locating the error step but incurs higher costs and risks accumulating errors.
Binary Search: A compromise between the first two methods, this strategy repeatedly divides the log in half, using the LLM to determine which segment contains the error. It then recursively searches the identified segment, offering a balance of cost and performance.
 
Experimental Results and Key Findings

Experiments were conducted in two settings: one where the LLM knows the ground truth answer to the problem the Multi-Agent system is trying to solve (With Ground Truth) and one where it does not (Without Ground Truth). The primary model used was GPT-4o, though other models were also tested. The systematic evaluation of these methods on the Who&When dataset yielded several important insights:

• A Long Way to Go: Current methods are far from perfect. Even the best-performing single method achieved an accuracy of only about 53.5% in identifying the responsible agent and a mere 14.2% in pinpointing the exact error step. Some methods performed even worse than random guessing, underscoring the difficulty of the task.
• No “All-in-One” Solution: Different methods excel at different aspects of the problem. The All-at-Once method is better at identifying “Who,” while the Step-by-Step method is more effective at determining “When.” The Binary Search method provides a middle-ground performance.

• Hybrid Approaches Show Promise but at a High Cost: The researchers found that combining different methods, such as using the All-at-Once approach to identify a potential agent and then applying the Step-by-Step method to find the error, can improve overall performance. However, this comes with a significant increase in computational cost.

• State-of-the-Art Models Struggle: Surprisingly, even the most advanced reasoning m…

  Konten dipersingkat otomatis.

  🔗 Sumber: syncedreview.com

  ---