Science Robotics最新文献_第5页

Hierarchically depicting vehicle trajectory with stability in complex environments 复杂环境下具有稳定性的车辆轨迹分层描绘

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-06-18 DOI: 10.1126/scirobotics.ads4551

Zhichao Han, Mengze Tian, Zaitian Gongye, Donglai Xue, Jiaxi Xing, Qianhao Wang, Yuman Gao, Jingping Wang, Chao Xu, Fei Gao

{"title":"Hierarchically depicting vehicle trajectory with stability in complex environments","authors":"Zhichao Han, Mengze Tian, Zaitian Gongye, Donglai Xue, Jiaxi Xing, Qianhao Wang, Yuman Gao, Jingping Wang, Chao Xu, Fei Gao","doi":"10.1126/scirobotics.ads4551","DOIUrl":"10.1126/scirobotics.ads4551","url":null,"abstract":"<div >The rapid development of autonomous robots has resulted in marked societal and economic benefits. However, enabling robots to navigate complex environments with human-like agility remains a formidable challenge. Unlike robots, humans excel at pathfinding because of their superior spatial awareness and their ability to leverage experience. Inspired by these observations, we designed a neural network to simulate the intuitive pathfinding abilities of humans, integrating global environmental information and previous experiences to identify feasible pathways. Experiments demonstrated that, unlike traditional algorithms whose efficiency deteriorates in complex settings, the proposed method maintains stable computational performance. To further enhance motion quality, we introduce a numerically stable spatiotemporal trajectory optimizer with a unique bilayer polynomial trajectory representation in flat space. This optimization leverages differential flatness to enhance efficiency and fundamentally eliminates singularities in the original problem, thereby robustly converging to continuous and feasible motion even in complex maneuvering scenarios. Our hierarchical motion planner, validated through large-scale maze experiments, combines front-end path planning with back-end trajectory refinement, achieving robust and efficient navigation. We anticipate that our planner will advance stable navigation for robots in complex environments, thereby propelling the progress of robotic autonomy.</div>","PeriodicalId":56029,"journal":{"name":"Science Robotics","volume":"10 103","pages":""},"PeriodicalIF":26.1,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144315435","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 compact neuromorphic system for ultra–energy-efficient, on-device robot localization 一个紧凑的神经形态系统，用于超节能，设备上的机器人定位

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-06-18 DOI: 10.1126/scirobotics.ads3968

Adam D. Hines, Michael Milford, Tobias Fischer

{"title":"A compact neuromorphic system for ultra–energy-efficient, on-device robot localization","authors":"Adam D. Hines, Michael Milford, Tobias Fischer","doi":"10.1126/scirobotics.ads3968","DOIUrl":"10.1126/scirobotics.ads3968","url":null,"abstract":"<div >Neuromorphic computing offers a transformative pathway to overcome the computational and energy challenges faced in deploying robotic localization and navigation systems at the edge. Visual place recognition, a critical component for navigation, is often hampered by the high resource demands of conventional systems, making them unsuitable for small-scale robotic platforms, which still require accurate long-endurance localization. Although neuromorphic approaches offer potential for greater efficiency, real-time edge deployment remains constrained by the complexity of biorealistic networks. To overcome this challenge, fusion of hardware and algorithms is critical when using this specialized computing paradigm. Here, we demonstrate a neuromorphic localization system that performs competitive place recognition in up to 8 kilometers of traversal using models as small as 180 kilobytes with 44,000 parameters while consuming less than 8% of the energy required by conventional methods. Our system, locational encoding with neuromorphic systems (LENS), integrates spiking neural networks, an event-based dynamic vision sensor, and a neuromorphic processor within a single SynSense Speck chip, enabling real-time, energy-efficient localization on a hexapod robot. When compared with a benchmark place recognition method, sum of absolute differences, LENS performs comparably in overall precision. LENS represents an accurate fully neuromorphic localization system capable of large-scale, on-device deployment for energy-efficient robotic place recognition. Neuromorphic computing enables resource-constrained robots to perform energy-efficient, accurate localization.</div>","PeriodicalId":56029,"journal":{"name":"Science Robotics","volume":"10 103","pages":""},"PeriodicalIF":26.1,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144315424","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

Multimodal information structuring with single-layer soft skins and high-density electrical impedance tomography 基于单层软皮和高密度电阻抗层析成像的多模态信息结构

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-06-11 DOI: 10.1126/scirobotics.adq2303

David Hardman, Thomas George Thuruthel, Fumiya Iida

{"title":"Multimodal information structuring with single-layer soft skins and high-density electrical impedance tomography","authors":"David Hardman, Thomas George Thuruthel, Fumiya Iida","doi":"10.1126/scirobotics.adq2303","DOIUrl":"10.1126/scirobotics.adq2303","url":null,"abstract":"<div >The human skin can reliably capture a wide range of multimodal data over a large surface while providing a soft interface. Artificial technologies using microelectromechanical systems (MEMS) can emulate these biological functions but present numerous challenges in fabrication, delamination due to soft-rigid interfaces, and electrical interference. To address these difficulties, we present a single-layer multimodal sensory skin made using only a highly sensitive hydrogel membrane. Using electrical impedance tomography techniques, we accessed up to 863,040 conductive pathways across the membrane, allowing us to identify at least six distinct types of multimodal stimuli, including human touch, damage, multipoint insulated presses, and local heating. Through comprehensive physical testing, we demonstrate that the highly redundant and coupled sensory information from these pathways can be structured using data-driven techniques, selecting which pathways should be monitored for efficient multimodal perception. To demonstrate our approach’s versatility, we cast the hydrogel into the shape and size of an adult human hand. Using our information structuring strategy, we demonstrate the hand’s ability to predict environmental conditions, localize human touch, and generate proprioceptive data. Our framework addresses the challenge of physically extracting meaningful information in multimodal soft sensing, opening new directions for the information-led design of single-layer skins in sensitive systems.</div>","PeriodicalId":56029,"journal":{"name":"Science Robotics","volume":"10 103","pages":""},"PeriodicalIF":26.1,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144260353","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

Roboticists are grappling with space debris 机器人专家正在努力处理太空碎片

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-06-11 DOI: 10.1126/scirobotics.adt5685

Jake J. Abbott

引用次数: 0

In situ foliar augmentation of multiple species for optical phenotyping and bioengineering using soft robotics 利用软机器人技术进行多物种的原位叶面扩增，用于光学表型分析和生物工程

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-06-11 DOI: 10.1126/scirobotics.adu2394

Mehmet Mert İlman, Annika Huber, Anand K. Mishra, Sabyasachi Sen, Fumin Wang, Tiffany Lin, Georg Jander, Abraham D. Stroock, Robert F. Shepherd

{"title":"In situ foliar augmentation of multiple species for optical phenotyping and bioengineering using soft robotics","authors":"Mehmet Mert İlman, Annika Huber, Anand K. Mishra, Sabyasachi Sen, Fumin Wang, Tiffany Lin, Georg Jander, Abraham D. Stroock, Robert F. Shepherd","doi":"10.1126/scirobotics.adu2394","DOIUrl":"10.1126/scirobotics.adu2394","url":null,"abstract":"<div >Precision agriculture aims to increase crop yield while reducing the use of harmful chemicals, such as pesticides and excess fertilizer, using minimal, tailored interventions. However, these strategies are limited by factors such as sensor quality, which typically relies on visual plant expression, and the manual, destructive nature of many nonvisual measurement methods, including the Scholander pressure bomb. By automating more intimate interactions with foliage in vivo, it would be possible to inject chemical and biological probes that reveal more phenotypes—such as water stress in response to varying environmental conditions and visible gene expression to measure the success of gene engineering applications. To address this, we developed a soft robotic leaf gripper and stamping-injection method to improve foliar delivery of nanoscale synthetic and biological probes. This allows for nondestructive, in situ, multispecies applications. We used two probes: <i>Agrobacterium tumefaciens</i> carrying the <i>RUBY</i> gene as a reporter system for plant transformation and nanoparticle hydrogels for measuring leaf water potential (ψ). Our hourglass-shaped design enabled the gripper to exert higher forces with reduced radial expansion compared with conventional designs, achieving an injection success rate above 91%. Studies on sunflower (<i>Helianthus annuus</i> L.) and cotton (<i>Gossypium hirsutum</i> L.) showed that our method achieved an average 12-fold increase in infiltration areas, with substantially less leaf damage—3.6% in sunflower and none in cotton—compared with the needle-free syringe method. Enabling long periods of successful in vivo phenotyping on both species after precise and safe foliar delivery underscores the potential of the leaf gripper for robotic plant bioengineering.</div>","PeriodicalId":56029,"journal":{"name":"Science Robotics","volume":"10 103","pages":""},"PeriodicalIF":26.1,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.science.org/doi/reader/10.1126/scirobotics.adu2394","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144260352","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

High-speed control and navigation for quadrupedal robots on complex and discrete terrain 复杂离散地形下四足机器人的高速控制与导航

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-05-28 DOI: 10.1126/scirobotics.ads6192

Hyeongjun Kim, Hyunsik Oh, Jeongsoo Park, Yunho Kim, Donghoon Youm, Moonkyu Jung, Minho Lee, Jemin Hwangbo

{"title":"High-speed control and navigation for quadrupedal robots on complex and discrete terrain","authors":"Hyeongjun Kim, Hyunsik Oh, Jeongsoo Park, Yunho Kim, Donghoon Youm, Moonkyu Jung, Minho Lee, Jemin Hwangbo","doi":"10.1126/scirobotics.ads6192","DOIUrl":"10.1126/scirobotics.ads6192","url":null,"abstract":"<div >High-speed legged navigation in discrete and geometrically complex environments is a challenging task because of the high–degree-of-freedom dynamics and long-horizon, nonconvex nature of the optimization problem. In this work, we propose a hierarchical navigation pipeline for legged robots that can traverse such environments at high speed. The proposed pipeline consists of a planner and tracker module. The planner module finds physically feasible foothold plans by sampling-based optimization with fast sequential filtering using heuristics and a neural network. Subsequently, rollouts are performed in a physics simulation to identify the best foothold plan regarding the engineered cost function and to confirm its physical consistency. This hierarchical planning module is computationally efficient and physically accurate at the same time. The tracker aims to accurately step on the target footholds from the planning module. During the training stage, the foothold target distribution is given by a generative model that is trained competitively with the tracker. This process ensures that the tracker is trained in an environment with the desired difficulty. The resulting tracker can overcome terrains that are more difficult than what the previous methods could manage. We demonstrated our approach using Raibo, our in-house dynamic quadruped robot. The results were dynamic and agile motions: Raibo is capable of running on vertical walls, jumping a 1.3-meter gap, running over stepping stones at 4 meters per second, and autonomously navigating on terrains full of 30° ramps, stairs, and boxes of various sizes.</div>","PeriodicalId":56029,"journal":{"name":"Science Robotics","volume":"10 102","pages":""},"PeriodicalIF":26.1,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144165427","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

The multifaceted approach to embodied intelligence in robotics. 机器人中具身智能的多面方法。

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-05-28 DOI: 10.1126/scirobotics.adx2731

Cecilia Laschi

引用次数: 0

Advancing physical intelligence for autonomous soft robots 推进自主软体机器人的物理智能

IF 25 1区计算机科学

Science Robotics Pub Date : 2025-05-28 DOI: 10.1126/scirobotics.ads1292

Chi Chen, Pengju Shi, Zixiao Liu, Sidi Duan, Muqing Si, Chuanwei Zhang, Yingjie Du, Yichen Yan, Timothy J. White, Rebecca Kramer-Bottiglio, Metin Sitti, Tetsuya Iwasaki, Ximin He

引用次数: 0

Learning coordinated badminton skills for legged manipulators 学习腿式操纵器的羽毛球协调技巧

IF 26.1 1区计算机科学

Science Robotics Pub Date : 2025-05-28 DOI: 10.1126/scirobotics.adu3922

Yuntao Ma, Andrei Cramariuc, Farbod Farshidian, Marco Hutter

{"title":"Learning coordinated badminton skills for legged manipulators","authors":"Yuntao Ma, Andrei Cramariuc, Farbod Farshidian, Marco Hutter","doi":"10.1126/scirobotics.adu3922","DOIUrl":"10.1126/scirobotics.adu3922","url":null,"abstract":"<div >Coordinating the motion between lower and upper limbs and aligning limb control with perception are substantial challenges in robotics, particularly in dynamic environments. To this end, we introduce an approach for enabling legged mobile manipulators to play badminton, a task that requires precise coordination of perception, locomotion, and arm swinging. We propose a unified reinforcement learning–based control policy for whole-body visuomotor skills involving all degrees of freedom to achieve effective shuttlecock tracking and striking. This policy is informed by a perception noise model that uses real-world camera data, allowing for consistent perception error levels between simulation and deployment and encouraging learned active perception behaviors. Our method includes a shuttlecock prediction model and constrained reinforcement learning for robust motion control to enhance deployment readiness. Extensive experimental results in a variety of environments validate the robot’s capability to predict shuttlecock trajectories, navigate the service area effectively, and execute precise strikes against human players, demonstrating the feasibility of using legged mobile manipulators in complex and dynamic sports scenarios.</div>","PeriodicalId":56029,"journal":{"name":"Science Robotics","volume":"10 102","pages":""},"PeriodicalIF":26.1,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.science.org/doi/reader/10.1126/scirobotics.adu3922","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144165425","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

Embodied aerial physical interaction: Combining body and brain for robust interaction with unstructured environments. 具体化的空中物理互动：结合身体和大脑与非结构化环境的强大互动。

IF 25 1区计算机科学

Science Robotics Pub Date : 2025-05-21 DOI: 10.1126/scirobotics.ads0200

Emanuele Aucone,Stefano Mintchev

引用次数: 0