24 papers · ranked by Valyu relevance
Julian Ibarz, Jie Tan, Chelsea Finn, Mrinal Kalakrishnan + 2 more
'Peter Pástor' 'Sergey Levine'] Deep reinforcement learning (RL) has emerged as a promising approach for autonomously acquiring complex behaviors from low level sensor observations. Although a large portion of deep RL research has focused on applications in video games and simulated control, which does not connect with…
Francesco Capuano, Caroline Pascal, Adil Zouitine, Thomas Wolf + 1 more
Robot learning is at an inflection point, driven by rapid advancements in machine learning and the growing availability of large-scale robotics data. This shift from classical, model-based methods to data-driven, learning-based paradigms is unlocking unprecedented capabilities in autonomous systems. This tutorial…
Jiang Hua, Liangcai Zeng, Gongfa Li, Zhaojie Ju + 1 more
Dexterous manipulation of the robot is an important part of realizing intelligence, but manipulators can only perform simple tasks such as sorting and packing in a structured environment. In view of the existing problem, this paper presents a state-of-the-art survey on an intelligent robot with the capability of…
Xuan Xiao, Jiahang Liu, Zhipeng Wang, Yanmin Zhou + 4 more
'Qian Cheng' 'Bin He' 'Shuo Jiang'] Abstract: The proliferation of Large Language Models (LLMs) has s fueled a shift in robot learning from automation towards general embodied Artificial Intelligence (AI). Adopting foundation models together with traditional learning methods to robot learning has increasingly gained…
Lisa Gutzeit, Alexander Fabisch, Marc Otto, Jan Hendrik Metzen + 3 more
'Jonas Hansen' 'Frank Kirchner' 'Elsa Andrea Kirchner'] We describe the BesMan learning platform which allows learning robotic manipulation behavior. It is a stand-alone solution which can be combined with different robotic systems and applications. Behavior that is adaptive to task changes and different target…
Amir Ramezani Dooraki, Deok-Jin Lee
In recent years, machine learning (and as a result artificial intelligence) has experienced considerable progress. As a result, robots in different shapes and with different purposes have found their ways into our everyday life. These robots, which have been developed with the goal of human companionship, are here to…
Jacqueline Heinerman, Evert Haasdijk, A. E. Eiben
Robot-to-robot learning, a specific case of social learning in robotics, enables multiple robots to share learned skills while completing a task. The literature offers various statements of its benefits. Robots using this type of social learning can reach a higher performance, an increased learning speed, or both…
Agnese Augello, Linda Daniela, Manuel Gentile, Dirk Ifenthaler + 1 more
'Giovanni Pilato'] Robots are increasingly being introduced in social environments to support the process of learning (e.g., Atmatzidou and Demetriadis, [2]; El Hamamsy et al., [11]; Kory-Westlund and Breazeal, [16]; Vogt et al., [27]) with different roles, such as smart teaching platforms, assistants, and in some…
Matthias Kerzel, Theresa Pekarek-Rosin, Erik Strahl, Stefan Heinrich + 1 more
'Stefan Wermter'] To overcome novel challenges in complex domestic environments, humanoid robots can learn from human teachers. We propose that the capability for social interaction should be a key factor in this teaching process and benefits both the subjective experience of the human user and the learning process…
Annalisa Taylor, Thomas A. Berrueta, Todd Murphey
Active learning is a decision-making process. In both abstract and physical settings, active learning demands both analysis and action. This is a review of active learning in robotics, focusing on methods amenable to the demands of embodied learning systems. Robots must be able to learn efficiently and flexibly through…
Sanaz Bazaz Behbahani, Siddharth R. Chhatpar, Said Zahrai, Vishakh Duggal + 1 more
'Vishakh Duggal' 'Mohak Sukhwani'] Abstract—Machine learning, artificial intelligence and especially deep learning based approaches are often used to simplify or eliminate the burden of programming industrial robots. Using these approaches robots inherently learn a skill instead of being programmed using strict and…
Alexander Fabisch, Christoph Petzoldt, Marc Otto, Frank Kirchner
Recent success of machine learning in many domains has been overwhelming, which often leads to false expectations regarding the capabilities of behavior learning in robotics. In this survey, we analyze the current state of machine learning for robotic behaviors. We will give a broad overview of behaviors that have been…
Naoto Yoshida, Hoshinori Kanazawa, Yasuo Kuniyoshi
Homeostasis is a fundamental property for the survival of animals. Computational reinforcement learning provides a theoretically sound framework for learning autonomous agents. However, the definition of a unified motivational signal (i.e., reward) for integrated survival behaviours has been largely underexplored.…
Raha Kannan, Maribel Gendreau, Alex Hatch, Sydney K. Free + 5 more
As the relevance of neuroscience in education grows, effective methods for teaching this complex subject in high school classrooms remain elusive. Integrating classroom experiments with brain-based robots offers a promising solution. This paper presents a structured curriculum designed around the use of camera-equipped…
Jongmin M. Lee, Temesgen Gebrekristos, Dalia De Santis, Mahdieh Nejati-Javaremi + 4 more
Learning to perform everyday tasks, using a complex robot, presents a nested problem. It is nested, because, on the surface, there is a problem of robot control—but within it, there lies a deeper, more challenging problem that demands the control nuances necessary to perform complicated functional tasks. For…
Joseph W. Barter, Henry H. Yin
Terrestrial locomotion presents tremendous computational challenges on account of the enormous degrees of freedom in legged animals, and the complex and unpredictable properties of the natural environment and the effectors. Yet the nervous system can achieve locomotion with ease. Here we introduce a quadrupedal robot…
Authors not listed
Machine olfaction—the artificial replication of the sense of smell—faces significant challenges due to the absence of large, standardized training datasets. Unlike vision, language, and audio models, which benefit from extensive corpora such as ImageNet, GLUE, and AudioSet, olfaction lacks scaled equivalents and…
Christopher A. Harris, Stanislav Mircic, Zachary Reining, Marcio Amorim + 4 more
Understanding the brain is a fascinating challenge, captivating the scientific community and the public alike. The lack of effective treatment for most brain disorders makes the training of the next generation of neuroscientists, engineers and physicians a key concern. Over the past decade there has been a growing…
Sivakumar Balasubramanian, Sandeep Guguloth, Javeed Shaikh Mohammed, S. Sujatha
Current evidence indicates that individual joint training with robotic devices can be as effective as multi-joint training for the arm. This makes a case for developing simpler and more compact robots for training individual joints of the arm. Such robots have the highest potential for clinical translation. To this…
Charlotte Canteloup, Joonho Lee, Samuel Zimmermann, Morgane Alvino + 3 more
Animal-robot interaction studies have been of increasing interest in research, but most of these studies have involved robots interacting with insects, birds, and frogs in laboratory settings. To date, only two studies used non-human primates and no behavioral study has tested the social integration of a robot in a…
Alexander Pomberger, Nicholas Jose, David Walz, Jens Meissner + 4 more
Buffer solutions have tremendous importance in biological systems and in formulated products. Whilst the pH response upon acid/base addition to a mixture containing a single buffer can be described by the Henderson-Hasselbalch equation, modelling the pH response for multi-buffered poly-protic systems after acid/base…
Zhichu Ren, Zhen Zhang, Yunsheng Tian, Ju Li
Autonomous laboratories were previously controlled mainly by scripting languages such as Python, limiting their usage among experimentalists. The recent release of OpenAI's ChatGPT API's function calling feature has enabled seamless integration and execution of Python subroutines in experimental workflows using voice…
Authors not listed
Process chemistry creates scalable routes for new lead molecules and is a crucial but laborious stage in pharmaceutical and agrochemical development cycles. We have built an automated process chemistry platform that tackles late-stage process development. The modular workflow integrates both industry-standard tools and…
Naruki Yoshikawa, Kourosh Darvish, Animesh Garg, Alan Aspuru-Guzik
Self-driving laboratories promise to democratize automated chemical laboratories. Accurate liquid handling is an essential operation in the context of chemical labs, and consequently a self-driving laboratory will require a robotic liquid handling and transfer. Although many pipettes are available for human scientists…