Category Ranking
98%
Total Visits
921
Avg Visit Duration
2 minutes
Citations
20
Article Abstract
In this paper, we propose an active touch sensing algorithm designed for robust hole localization in 3D objects, specifically aimed at assembly tasks such as peg-in-hole operations. Unlike general object detection algorithms, our solution is tailored for precise localization of features like hole openings using sparse tactile feedback. The method builds on a prior 3D map of the object and employs a series of iterative search algorithms to refine localization by aligning tactile sensing data with the object's shape. It is specifically designed for objects composed of multiple parallel surfaces located at distinct heights; a common characteristic in many assembly tasks. In addition to the deterministic approach, we introduce a probabilistic version of the algorithm, which effectively compensates for sensor noise and inaccuracies in the 3D map. This probabilistic framework significantly improves the algorithm's resilience in real-world environments, ensuring reliable performance even under imperfect conditions. We validate the method's effectiveness for several assembly tasks, such as inserting a plug into a socket, demonstrating its speed and accuracy. The proposed algorithm outperforms traditional search strategies, offering a robust solution for assembly operations in industrial and domestic applications with limited sensory input.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC12349164 | PMC |
| http://dx.doi.org/10.3390/s25154567 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Using reinforcement learning in genome assembly: in-depth analysis of a Q-learning assembler.
Front Bioinform
August 2025
Data Science, Vale Institute of Technology, Belém, Pará, Brazil.
Genome assembly remains an unsolved problem, and de novo strategies (i.e., those run without a reference) are relevant but computationally complex tasks in genomics.
View Article and Find Full Text PDFLarge Language Models (LLMs), AI agents and co-scientists promise to accelerate scientific discovery across fields ranging from chemistry to biology. Bioinformatics- the analysis of DNA, RNA and protein sequences plays a crucial role in biological research and is especially amenable to AI-driven automation given its computational nature. Here, we assess the bioinformatics capabilities of three popular general-purpose LLMs on a set of tasks covering basic analytical questions that include code writing and multi-step reasoning in the domain.
View Article and Find Full Text PDFMicroswimmers That Flex: Advancing Microswimmers with Templated Assembly and Responsive DNA Nanostructures.
Acc Mater Res
August 2025
Department of Mechanical Engineering, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, Pennsylvania 15213, United States.
The concept of micrometer-scale swimming robots, also known as microswimmers, navigating the human body to perform robotic tasks has captured the public imagination and inspired researchers through its numerous representations in popular media. This attention highlights the enormous interest in and potential of this technology for biomedical applications, such as cargo delivery, diagnostics, and minimally invasive surgery, as well as for applications in microfluidics and manufacturing. To achieve the collective behavior and control required for microswimmers to effectively perform such actions within complex, in vivo and microfluidic environments, they must meet a strict set of engineering criteria.
View Article and Find Full Text PDFModel Quality Assessment for CASP16.
Proteins
August 2025
Cambridge Institute for Medical Research, University of Cambridge, Cambridge, UK.
The CASP16 evaluation of model accuracy (EMA) experiment assessed the ability of predictors to estimate the accuracy of predicted models, with a particular emphasis on multimeric assemblies. Expanding on the CASP15 framework, CASP16 introduced a new evaluation mode (QMODE3) focused on selecting high-quality models from large-scale AlphaFold2-derived model pools generated by MassiveFold. Three primary evaluation tasks were therefore conducted: QMODE1 assessed global structure accuracy, QMODE2 focused on the accuracy of interface residues, and QMODE3 tested model selection performance.
View Article and Find Full Text PDFPrecise and dexterous robotic manipulation via human-in-the-loop reinforcement learning.
Sci Robot
August 2025
Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, CA 94720, USA.
Article Synopsis
- Robotic manipulation is a key challenge in robotics, with current methods facing limitations in design, performance, and data needs, affecting their real-world application.
- Reinforcement learning (RL) offers an alternative by allowing robots to learn complex skills through interaction, but challenges like sample efficiency and safety remain.
- Our human-in-the-loop, vision-based RL system has achieved significant improvements in dexterous manipulation tasks in the real world, doubling task success rates and increasing execution speed within just a couple of hours of training.