Science Robotics最新文献_第5页

Safety-assured high-speed navigation for MAVs

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-01-29 DOI: 10.1126/scirobotics.ado6187

Yunfan Ren, Fangcheng Zhu, Guozheng Lu, Yixi Cai, Longji Yin, Fanze Kong, Jiarong Lin, Nan Chen, Fu Zhang

{"title":"Safety-assured high-speed navigation for MAVs","authors":"Yunfan Ren, Fangcheng Zhu, Guozheng Lu, Yixi Cai, Longji Yin, Fanze Kong, Jiarong Lin, Nan Chen, Fu Zhang","doi":"10.1126/scirobotics.ado6187","DOIUrl":"10.1126/scirobotics.ado6187","url":null,"abstract":"<div >Micro air vehicles (MAVs) capable of high-speed autonomous navigation in unknown environments have the potential to improve applications like search and rescue and disaster relief, where timely and safe navigation is critical. However, achieving autonomous, safe, and high-speed MAV navigation faces systematic challenges, necessitating reduced vehicle weight and size for high-speed maneuvering, strong sensing capability for detecting obstacles at a distance, and advanced planning and control algorithms maximizing flight speed while ensuring obstacle avoidance. Here, we present the safety-assured high-speed aerial robot (SUPER), a compact MAV with a 280-millimeter wheelbase and a thrust-to-weight ratio greater than 5.0, enabling agile flight in cluttered environments. SUPER uses a lightweight three-dimensional light detection and ranging (LIDAR) sensor for accurate, long-range obstacle detection. To ensure high-speed flight while maintaining safety, we introduced an efficient planning framework that directly plans trajectories using LIDAR point clouds. In each replanning cycle, two trajectories were generated: one in known free spaces to ensure safety and another in both known and unknown spaces to maximize speed. Compared with baseline methods, this framework reduced failure rates by 35.9 times while flying faster and with half the planning time. In real-world tests, SUPER achieved autonomous flights at speeds exceeding 20 meters per second, successfully avoiding thin obstacles and navigating narrow spaces. SUPER represents a milestone in autonomous MAV systems, bridging the gap from laboratory research to real-world applications.</div>","PeriodicalId":56029,"journal":{"name":"Science Robotics","volume":"10 98","pages":""},"PeriodicalIF":26.1,"publicationDate":"2025-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.science.org/doi/reader/10.1126/scirobotics.ado6187","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143056756","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Humanoid robot learning of complex behaviors with LLMs

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-01-22 DOI: 10.1126/scirobotics.adv4627

Amos Matsiko

引用次数: 0

Development of compositionality through interactive learning of language and action of robots 通过语言和机器人动作的互动学习来发展组合性

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-01-22 DOI: 10.1126/scirobotics.adp0751

Prasanna Vijayaraghavan, Jeffrey Frederic Queißer, Sergio Verduzco Flores, Jun Tani

{"title":"Development of compositionality through interactive learning of language and action of robots","authors":"Prasanna Vijayaraghavan, Jeffrey Frederic Queißer, Sergio Verduzco Flores, Jun Tani","doi":"10.1126/scirobotics.adp0751","DOIUrl":"10.1126/scirobotics.adp0751","url":null,"abstract":"<div >Humans excel at applying learned behavior to unlearned situations. A crucial component of this generalization behavior is our ability to compose/decompose a whole into reusable parts, an attribute known as compositionality. One of the fundamental questions in robotics concerns this characteristic: How can linguistic compositionality be developed concomitantly with sensorimotor skills through associative learning, particularly when individuals only learn partial linguistic compositions and their corresponding sensorimotor patterns? To address this question, we propose a brain-inspired neural network model that integrates vision, proprioception, and language into a framework of predictive coding and active inference on the basis of the free-energy principle. The effectiveness and capabilities of this model were assessed through various simulation experiments conducted with a robot arm. Our results show that generalization in learning to unlearned verb-noun compositions is significantly enhanced when training variations of task composition are increased. We attribute this to self-organized compositional structures in linguistic latent state space being influenced substantially by sensorimotor learning. Ablation studies show that visual attention and working memory are essential to accurately generate visuomotor sequences to achieve linguistically represented goals. These insights advance our understanding of mechanisms underlying development of compositionality through interactions of linguistic and sensorimotor experience.</div>","PeriodicalId":56029,"journal":{"name":"Science Robotics","volume":"10 98","pages":""},"PeriodicalIF":26.1,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143020311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A neuromechanics solution for adjustable robot compliance and accuracy 可调节机器人顺应性和精度的神经力学解决方案

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-01-22 DOI: 10.1126/scirobotics.adp2356

Ignacio Abadía, Alice Bruel, Grégoire Courtine, Auke J. Ijspeert, Eduardo Ros, Niceto R. Luque

{"title":"A neuromechanics solution for adjustable robot compliance and accuracy","authors":"Ignacio Abadía, Alice Bruel, Grégoire Courtine, Auke J. Ijspeert, Eduardo Ros, Niceto R. Luque","doi":"10.1126/scirobotics.adp2356","DOIUrl":"10.1126/scirobotics.adp2356","url":null,"abstract":"<div >Robots have to adjust their motor behavior to changing environments and variable task requirements to successfully operate in the real world and physically interact with humans. Thus, robotics strives to enable a broad spectrum of adjustable motor behavior, aiming to mimic the human ability to function in unstructured scenarios. In humans, motor behavior arises from the integrative action of the central nervous system and body biomechanics; motion must be understood from a neuromechanics perspective. Nervous regions such as the cerebellum facilitate learning, adaptation, and coordination of our motor responses, ultimately driven by muscle activation. Muscles, in turn, self-stabilize motion through mechanical viscoelasticity. In addition, the agonist-antagonist arrangement of muscles surrounding joints enables cocontraction, which can be regulated to enhance motion accuracy and adapt joint stiffness, thereby providing impedance modulation and broadening the motor repertoire. Here, we propose a control solution that harnesses neuromechanics to enable adjustable robot motor behavior. Our solution integrates a muscle model that replicates mechanical viscoelasticity and cocontraction together with a cerebellar network providing motor adaptation. The resulting cerebello-muscular controller drives the robot through torque commands in a feedback control loop. Changes in cocontraction modify the muscle dynamics, and the cerebellum provides motor adaptation without relying on prior analytical solutions, driving the robot in different motor tasks, including payload perturbations and operation across unknown terrains. Experimental results show that cocontraction modulates robot stiffness, performance accuracy, and robustness against external perturbations. Through cocontraction modulation, our cerebello-muscular torque controller enables a broad spectrum of robot motor behavior.</div>","PeriodicalId":56029,"journal":{"name":"Science Robotics","volume":"10 98","pages":""},"PeriodicalIF":26.1,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143020312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Would a robot ever get angry enough to attack a person?

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-01-22 DOI: 10.1126/scirobotics.adv3128

Robin R. Murphy

引用次数: 0

Surmounting the ceiling effect of motor expertise by novel sensory experience with a hand exoskeleton 通过手外骨骼的新颖感官体验，超越了运动专业知识的天花板效应

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-01-15 DOI: 10.1126/scirobotics.adn3802

Shinichi Furuya, Takanori Oku, Hayato Nishioka, Masato Hirano

{"title":"Surmounting the ceiling effect of motor expertise by novel sensory experience with a hand exoskeleton","authors":"Shinichi Furuya, Takanori Oku, Hayato Nishioka, Masato Hirano","doi":"10.1126/scirobotics.adn3802","DOIUrl":"10.1126/scirobotics.adn3802","url":null,"abstract":"<div >For trained individuals such as athletes and musicians, learning often plateaus after extensive training, known as the “ceiling effect.” One bottleneck to overcome it is having no prior physical experience with the skill to be learned. Here, we challenge this issue by exposing expert pianists to fast and complex finger movements that cannot be performed voluntarily, using a hand exoskeleton robot that can move individual fingers quickly and independently. Although the skill of moving the fingers quickly plateaued through weeks of piano practice, passive exposure to otherwise impossible complex finger movements generated by the exoskeleton robot at a speed faster than the pianists’ fastest one enabled them to play faster. Neither a training for fast but simple finger movements nor one for slow but complex movements with the exoskeleton enhanced the overtrained motor skill. The exoskeleton training with one hand also improved the motor skill of the untrained contralateral hand, demonstrating the intermanual transfer effect. The training altered patterns of coordinated activities across multiple finger muscles during piano playing but not in general motor and somatosensory functions or in anatomical characteristics of the hand (range of motion). Patterns of the multifinger movements evoked by transcranial magnetic stimulation over the left motor cortex were also changed through passive exposure to fast and complex finger movements, which accompanied increased involvement of constituent movement elements characterizing the individuated finger movements. The results demonstrate evidence that somatosensory exposure to an unexperienced motor skill allows surmounting of the ceiling effect in a task-specific but effector-independent manner.</div>","PeriodicalId":56029,"journal":{"name":"Science Robotics","volume":"10 98","pages":""},"PeriodicalIF":26.1,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.science.org/doi/reader/10.1126/scirobotics.adn3802","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142986388","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Merging motoneuron and postural synergies in prosthetic hand design for natural bionic interfacing 融合运动神经元和姿势协同作用的假手自然仿生界面设计

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-01-15 DOI: 10.1126/scirobotics.ado9509

Patricia Capsi-Morales, Deren Y. Barsakcioglu, Manuel G. Catalano, Giorgio Grioli, Antonio Bicchi, Dario Farina

{"title":"Merging motoneuron and postural synergies in prosthetic hand design for natural bionic interfacing","authors":"Patricia Capsi-Morales, Deren Y. Barsakcioglu, Manuel G. Catalano, Giorgio Grioli, Antonio Bicchi, Dario Farina","doi":"10.1126/scirobotics.ado9509","DOIUrl":"10.1126/scirobotics.ado9509","url":null,"abstract":"<div >Despite the advances in bionic reconstruction of missing limbs, the control of robotic limbs is still limited and, in most cases, not felt to be as natural by users. In this study, we introduce a control approach that combines robotic design based on postural synergies and neural decoding of synergistic behavior of spinal motoneurons. We developed a soft prosthetic hand with two degrees of actuation that realizes postures in a two-dimensional linear manifold generated by two postural synergies. Through a manipulation task in nine participants without physical impairment, we investigated how to map neural commands to the postural synergies. We found that neural synergies outperformed classic muscle synergies in terms of dimensionality and robustness. Leveraging these findings, we developed an online method to map the decoded neural synergies into continuous control of the two-synergy prosthetic hand, which was tested on 11 participants without physical impairment and three prosthesis users in real-time scenarios. Results demonstrated that combined neural and postural synergies allowed accurate and natural control of coordinated multidigit actions (>90% of the continuous mechanical manifold could be reached). The target hit rate for specific hand postures was higher with neural synergies compared with muscle synergies, with the difference being particularly pronounced for prosthesis users (prosthesis users, 82.5% versus 35.0%; other participants, 79.5% versus 54.5%). This demonstration of codesign of multisynergistic robotic hands and neural decoding algorithms enabled users to achieve natural modular control to span infinite postures across a two-dimensional space and to execute dexterous tasks, including in-hand manipulation, not feasible with other approaches.</div>","PeriodicalId":56029,"journal":{"name":"Science Robotics","volume":"10 98","pages":""},"PeriodicalIF":26.1,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142986812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Acrobatics at the insect scale: A durable, precise, and agile micro–aerial robot 昆虫级杂技：一种耐用、精确、敏捷的微型空中机器人

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-01-15 DOI: 10.1126/scirobotics.adp4256

Suhan Kim, Yi-Hsuan Hsiao, Zhijian Ren, Jiashu Huang, Yufeng Chen

{"title":"Acrobatics at the insect scale: A durable, precise, and agile micro–aerial robot","authors":"Suhan Kim, Yi-Hsuan Hsiao, Zhijian Ren, Jiashu Huang, Yufeng Chen","doi":"10.1126/scirobotics.adp4256","DOIUrl":"10.1126/scirobotics.adp4256","url":null,"abstract":"<div >Aerial insects are exceptionally agile and precise owing to their small size and fast neuromotor control. They perform impressive acrobatic maneuvers when evading predators, recovering from wind gust, or landing on moving objects. Flapping-wing propulsion is advantageous for flight agility because it can generate large changes in instantaneous forces and torques. During flapping-wing flight, wings, hinges, and tendons of pterygote insects endure large deformation and high stress hundreds of times each second, highlighting the outstanding flexibility and fatigue resistance of biological structures and materials. In comparison, engineered materials and microscale structures in subgram micro–aerial vehicles (MAVs) exhibit substantially shorter lifespans. Consequently, most subgram MAVs are limited to hovering for less than 10 seconds or following simple trajectories at slow speeds. Here, we developed a 750-milligram flapping-wing MAV that demonstrated substantially improved lifespan, speed, accuracy, and agility. With transmission and hinge designs that reduced off-axis torsional stress and deformation, the robot achieved a 1000-second hovering flight, two orders of magnitude longer than existing subgram MAVs. This robot also performed complex flight trajectories with under 1-centimeter root mean square error and more than 30 centimeters per second average speed. With a lift-to-weight ratio of 2.2 and a maximum ascending speed of 100 centimeters per second, this robot demonstrated double body flips at a rotational rate exceeding that of the fastest aerial insects and larger MAVs. These results highlight insect-like flight endurance, precision, and agility in an at-scale MAV, opening opportunities for future research on sensing and power autonomy.</div>","PeriodicalId":56029,"journal":{"name":"Science Robotics","volume":"10 98","pages":""},"PeriodicalIF":26.1,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142986810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Even teleoperated robots are discriminated against in science fictions. 即使是遥控机器人在科幻小说中也受到歧视。

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2024-12-18 DOI: 10.1126/scirobotics.adu6332

Robin R Murphy

引用次数: 0

Counterfactual rewards promote collective transport using individually controlled swarm microrobots 反事实奖励促进使用个体控制的群体微型机器人进行集体运输

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2024-12-18 DOI: 10.1126/scirobotics.ado5888

Veit-Lorenz Heuthe, Emanuele Panizon, Hongri Gu, Clemens Bechinger

{"title":"Counterfactual rewards promote collective transport using individually controlled swarm microrobots","authors":"Veit-Lorenz Heuthe, Emanuele Panizon, Hongri Gu, Clemens Bechinger","doi":"10.1126/scirobotics.ado5888","DOIUrl":"10.1126/scirobotics.ado5888","url":null,"abstract":"<div >Swarm robots offer fascinating opportunities to perform complex tasks beyond the capabilities of individual machines. Just as a swarm of ants collectively moves large objects, similar functions can emerge within a group of robots through individual strategies based on local sensing. However, realizing collective functions with individually controlled microrobots is particularly challenging because of their micrometer size, large number of degrees of freedom, strong thermal noise relative to the propulsion speed, and complex physical coupling between neighboring microrobots. Here, we implemented multiagent reinforcement learning (MARL) to generate a control strategy for up to 200 microrobots whose motions are individually controlled by laser spots. During the learning process, we used so-called counterfactual rewards that automatically assign credit to the individual microrobots, which allows fast and unbiased training. With the help of this efficient reward scheme, swarm microrobots learn to collectively transport a large cargo object to an arbitrary position and orientation, similar to ant swarms. We show that this flexible and versatile swarm robotic system is robust to variations in group size, the presence of malfunctioning units, and environmental noise. In addition, we let the robot swarms manipulate multiple objects simultaneously in a demonstration experiment, highlighting the benefits of distributed control and independent microrobot motion. Control strategies such as ours can potentially enable complex and automated assembly of mobile micromachines, programmable drug delivery capsules, and other advanced lab-on-a-chip applications.</div>","PeriodicalId":56029,"journal":{"name":"Science Robotics","volume":"9 97","pages":""},"PeriodicalIF":26.1,"publicationDate":"2024-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142841881","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0