Pedestrian trajectory prediction can significantly enhance the perception and decision-making capabilities of autonomous driving systems and intelligent surveillance systems based on camera sensors by predicting the s...Pedestrian trajectory prediction can significantly enhance the perception and decision-making capabilities of autonomous driving systems and intelligent surveillance systems based on camera sensors by predicting the states and behavior intentions of surrounding pedestrians.However,existing trajectory prediction methods remain failing to effectively model the diverse and complex interactions in the real world,including pedestrian-pedestrian interactions and pedestrian-environment interactions.Besides,these methods are not effective in capturing and characterizing the multimodal property of future trajectories.To address these challenges above,we propose to devise a handdesigned graph convolution and spatial cross attention to dynamically capture the diverse spatial interactions between pedestrians.To effectively explore the impact of scenarios on pedestrian trajectory,we build a pedestrian map,which can reflect the scene constraints and pedestrian motion preferences.Meanwhile,we construct a trajectory multimodality-aware module to capture the different potential mode implicit in diverse social behaviors for pedestrian future trajectory uncertainty.Finally,we compared the proposed method with trajectory prediction baselines on commonly used public pedestrian benchmarks,demonstrating the superior performance of our approach.展开更多
THE tremendous impact of large models represented by ChatGPT[1]-[3]makes it necessary to con-sider the practical applications of such models[4].However,for an artificial intelligence(AI)to truly evolve,it needs to pos...THE tremendous impact of large models represented by ChatGPT[1]-[3]makes it necessary to con-sider the practical applications of such models[4].However,for an artificial intelligence(AI)to truly evolve,it needs to possess a physical“body”to transition from the virtual world to the real world and evolve through interaction with the real environments.In this context,“embodied intelligence”has sparked a new wave of research and technology,leading AI beyond the digital realm into a new paradigm that can actively act and perceive in a physical environment through tangible entities such as robots and automated devices[5].展开更多
Origami and kirigami structures originate from the ancient art of paper folding and cutting.They offer the opportunity to achieve superior performances and unusual function-alities brought by complex three-dimensional...Origami and kirigami structures originate from the ancient art of paper folding and cutting.They offer the opportunity to achieve superior performances and unusual function-alities brought by complex three-dimensional geometries using approachable manufacturing technologies that work on two-dimensional sheets.展开更多
As a cross-cutting field between ocean development and multi-robot system(MRS),the underwater multi-robot system(UMRS)has gained increasing attention from researchers and engineers in recent decades.In this paper,we p...As a cross-cutting field between ocean development and multi-robot system(MRS),the underwater multi-robot system(UMRS)has gained increasing attention from researchers and engineers in recent decades.In this paper,we present a comprehensive survey of cooperation issues,one of the key components of UMRS,from the perspective of the emergence of new functions.More specifically,we categorize the cooperation in terms of task-space,motion-space,measurement-space,as well as their combination.Further,we analyze the architecture of UMRS from three aspects,i.e.,the performance of the individual underwater robot,the new functions of underwater robots,and the technical approaches of MRS.To conclude,we have discussed related promising directions for future research.This survey provides valuable insight into the reasonable utilization of UMRS to attain diverse underwater tasks in complex ocean application scenarios.展开更多
Dear Editor,Quadratic programming problems(QPs)receive a lot of attention in various fields of science computing and engineering applications,such as manipulator control[1].Recursive neural network(RNN)is considered t...Dear Editor,Quadratic programming problems(QPs)receive a lot of attention in various fields of science computing and engineering applications,such as manipulator control[1].Recursive neural network(RNN)is considered to be a powerful QPs solver due to its parallel processing capability and feasibility of hardware implementation[2].展开更多
Instability-induced wrinkle patterns of thin sheets are ubiquitous in nature,which often result in origami-like patterns that provide inspiration for the engineering of origami designs.Inspired by instability-induced ...Instability-induced wrinkle patterns of thin sheets are ubiquitous in nature,which often result in origami-like patterns that provide inspiration for the engineering of origami designs.Inspired by instability-induced origami patterns,we propose a computational origami design method based on the nonlinear analysis of loaded thin sheets and topology optimization.The bar-and-hinge model is employed for the nonlinear structural analysis,added with a displacement perturbation strategy to initiate out-of-plane buckling.Borrowing ideas from topology optimization,a continuous crease indicator is introduced as the design variable to indicate the state of a crease,which is penalized by power functions to establish the mapping relationships between the crease indicator and hinge properties.Minimizing the structural strain energy with a crease length constraint,we are able to evolve a thin sheet into an origami structure with an optimized crease pattern.Two examples with different initial setups are illustrated,demonstrating the effectiveness and feasibility of the method.展开更多
Polyelectrolyte(PE)gels,distinguished by their unique stimuli-responsive swelling behavior,serve as the basis of broad applications,such as artificial muscles and drug delivery.In this work,we present a theoretical mo...Polyelectrolyte(PE)gels,distinguished by their unique stimuli-responsive swelling behavior,serve as the basis of broad applications,such as artificial muscles and drug delivery.In this work,we present a theoretical model to analyze the electrostatics and its contribution to the swelling behavior of PE gels in salt solutions.By minimizing the free energy of PE gels,we obtain two distinct scaling regimes for the swelling ratio at equilibrium with respect to the salt concentration.We compare our predictions for the swelling ratio with experimental measurements,which show excellent agreement.In addition,we employ a finite element method to assess the applicability range of our theoretical model and assumptions.We anticipate that our model will also provide valuable insights into drug adsorption and release,deformation of red blood cells,4D printing and soft robotics,where the underlying mechanism of swelling remains enigmatic.展开更多
This paper presents a distributed scheme with limited communications, aiming to achieve cooperative motion control for multiple omnidirectional mobile manipulators(MOMMs).The proposed scheme extends the existing singl...This paper presents a distributed scheme with limited communications, aiming to achieve cooperative motion control for multiple omnidirectional mobile manipulators(MOMMs).The proposed scheme extends the existing single-agent motion control to cater to scenarios involving the cooperative operation of MOMMs. Specifically, squeeze-free cooperative load transportation is achieved for the end-effectors of MOMMs by incorporating cooperative repetitive motion planning(CRMP), while guiding each individual to desired poses. Then, the distributed scheme is formulated as a time-varying quadratic programming(QP) and solved online utilizing a noise-tolerant zeroing neural network(NTZNN). Theoretical analysis shows that the NTZNN model converges globally to the optimal solution of QP in the presence of noise. Finally, the effectiveness of the control design is demonstrated by numerical simulations and physical platform experiments.展开更多
Disordered hyperuniformity(DHU)is a recently discovered novel state of amorphous systems characterized by strongly suppressed density fluctuations at large length scales as in crystalline materials,which offers great ...Disordered hyperuniformity(DHU)is a recently discovered novel state of amorphous systems characterized by strongly suppressed density fluctuations at large length scales as in crystalline materials,which offers great potential for achieving unusual mechanical,electronic,and photonic properties.However,despite the fundamental and technological importance of thermal transport in amorphous solids,the effect of DHU remains largely unexplored.Here,we theoretically study thermal transport in a class of two-dimensional DHU materials—monolayer amorphous carbon(MAC).Beginning with a perfect graphene lattice,we continuously apply Stone-Wales transformations to generate a series of MAC models with varied degrees of disorder and defects,which are quantified through comprehensive structural analysis including the so-called hyperuniformity index(H),where a smaller H indicates a higher degree of hyperuniformity.Subsequently,we conduct molecular dynamics simulations to obtain the thermal conductivity(κ).A significant correlation between the thermal transport behavior and degree of hyperuniformity is observed,with the room-temperatureκdecreasing from 26.3 to 5.3 W m^(-1)K^(-1)while H is tuned from 0.0004 to 0.024.Remarkably,two distinct transport regimes are identified,including a nearly-DHU regime at small H(<0.005)whereκdrops sharply and a non-DHU region at larger H whereκbecomes relatively flat.Mode-resolved analysis reveals longer lifetime and higher participation ratio for the heat carriers in nearly-DHU MAC,implying that the hidden long-range correlations could support extended modes that enhance transport.Our work highlights the impact of DHU on the thermal properties of amorphous materials and represents a conceptual advancement that is worthy of future exploration.展开更多
Few-shot object detection receives much attention with the ability to detect novel class objects using limited annotated data.The transfer learning-based solution becomes popular due to its simple training with good a...Few-shot object detection receives much attention with the ability to detect novel class objects using limited annotated data.The transfer learning-based solution becomes popular due to its simple training with good accuracy,however,it is still challenging to enrich the feature diversity during the training process.And fine-grained features are also insufficient for novel class detection.To deal with the problems,this paper proposes a novel few-shot object detection method based on dual-domain feature fusion and patch-level attention.Upon original base domain,an elementary domain with more category-agnostic features is superposed to construct a two-stream backbone,which benefits to enrich the feature diversity.To better integrate various features,a dual-domain feature fusion is designed,where the feature pairs with the same size are complementarily fused to extract more discriminative features.Moreover,a patch-wise feature refinement termed as patch-level attention is presented to mine internal relations among the patches,which enhances the adaptability to novel classes.In addition,a weighted classification loss is given to assist the fine-tuning of the classifier by combining extra features from FPN of the base training model.In this way,the few-shot detection quality to novel class objects is improved.Experiments on PASCAL VOC and MS COCO datasets verify the effectiveness of the method.展开更多
The task of path planning in amphibious environments requires additional consideration due to the complexity of the amphibious environments.This paper presents a path planning method for an amphibious robot named\Amph...The task of path planning in amphibious environments requires additional consideration due to the complexity of the amphibious environments.This paper presents a path planning method for an amphibious robot named\AmphiRobot"with its dynamic constraints considered.First,an explicit dynamic model using Kane's method is presented.The hydrodynamic parameters are obtained through computational°uid dynamics simulations.Furthermore,a path planning method based on a hybrid¯reworks algorithm is proposed,combining the¯reworks algorithm and bare bones¯reworks algorithm,aiming at the amphibious robot's characteristics of multiple motion modes and working environments.The initially planned path is then smoothed using Dubins path under constraints determined by the dynamic model.Simulation reveals that the performance of the hybrid¯reworks algorithm approach is better than the¯reworks algorithm and bare bones¯reworks algorithm is applied separately in the amphibious environment scenarios.展开更多
Central pattern generators(CPGs)have been widely applied in robot motion control for the spontaneous output of coherent periodic rhythms.However,the underlying CPG network exhibits good convergence performance only wi...Central pattern generators(CPGs)have been widely applied in robot motion control for the spontaneous output of coherent periodic rhythms.However,the underlying CPG network exhibits good convergence performance only within a certain range of parameter spaces,and the coupling of oscillators affects the network output accuracy in complex topological relationships.Moreover,CPGs may diverge when parameters change drastically,and the divergence is irreversible,which is catastrophic for the control of robot motion.Therefore,normalized asymmetric CPGs(NA-CPGs)that normalize the amplitude parameters of Hopf-based CPGs and add a constraint function and a frequency regulation mechanism are proposed.NA-CPGs can realize parameter decoupling,precise amplitude output,and stable and rapid convergence,as well as asymmetric output waveforms.Thus,it can effectively cope with large parameter changes to avoid network oscillations and divergence.To optimize the parameters of the NA-CPG model,a reinforcement-learning-based online optimization method is further proposed.Meanwhile,a biomimetic robotic fish is illustrated to realize the whole optimization process.Simulations demonstrated that the designed NA-CPGs exhibit stable,secure,and accurate network outputs,and the proposed optimization method effectively improves the swimming speed and reduces the lateral swing of the multijoint robotic fish by 6.7%and 41.7%,respectively.The proposed approach provides a significant improvement in CPG research and can be widely employed in the field of robot motion control.展开更多
To address the challenging task of effective sound absorption in the low and broad frequency band for underwater structures,we propose a novel grating‐like anechoic layer by filling rubber blocks and an air backing l...To address the challenging task of effective sound absorption in the low and broad frequency band for underwater structures,we propose a novel grating‐like anechoic layer by filling rubber blocks and an air backing layer into metallic grating.The metallic gratings are incorporated into the anechoic layer as a skeleton for enhanced viscoelastic dissipation by promoting shear deformation between rubber and metal plates.The introduction of an air backing layer releases the bottom constraint of the rubber,thus intensifying its deformation under acoustic excitation.Based on the homogenization method and the transfer matrix method,a theoretical model is developed to evaluate the sound absorption performance of the proposed anechoic layer,which is validated against finite element simulation results.It is demonstrated that a sound absorption coefficient of the grating‐like anechoic layer of 0.8 can be achieved in the frequency range of 1294-10000 Hz.Given the importance of sound absorption at varying frequencies,the weighted average method is subsequently used to comprehensively evaluate the performance of the anechoic layer.Then,with structural density taken into consideration,an integrated index is proposed to further evaluate the acoustic properties of the proposed anechoic layer.Finally,the backing conditions and the boundary conditions of finite‐size structures are discussed.The results provide helpful theoretical guidance for designing novel acoustic metamaterials with broadband low‐frequency underwater sound absorption.展开更多
Grasp detection plays a critical role for robot manipulation.Mainstream pixel-wise grasp detection networks with encoder-decoder structure receive much attention due to good accuracy and efficiency.However,they usuall...Grasp detection plays a critical role for robot manipulation.Mainstream pixel-wise grasp detection networks with encoder-decoder structure receive much attention due to good accuracy and efficiency.However,they usually transmit the high-level feature in the encoder to the decoder,and low-level features are neglected.It is noted that low-level features contain abundant detail information,and how to fully exploit low-level features remains unsolved.Meanwhile,the channel information in high-level feature is also not well mined.Inevitably,the performance of grasp detection is degraded.To solve these problems,we propose a grasp detection network with hierarchical multi-scale feature fusion and inverted shuffle residual.Both low-level and high-level features in the encoder are firstly fused by the designed skip connections with attention module,and the fused information is then propagated to corresponding layers of the decoder for in-depth feature fusion.Such a hierarchical fusion guarantees the quality of grasp prediction.Furthermore,an inverted shuffle residual module is created,where the high-level feature from encoder is split in channel and the resultant split features are processed in their respective branches.By such differentiation processing,more high-dimensional channel information is kept,which enhances the representation ability of the network.Besides,an information enhancement module is added before the encoder to reinforce input information.The proposed method attains 98.9%and 97.8%in image-wise and object-wise accuracy on the Cornell grasping dataset,respectively,and the experimental results verify the effectiveness of the method.展开更多
To tackle the problem of aquatic environment pollution,a vision-based autonomous underwater garbage cleaning robot has been developed in our laboratory.We propose a garbage detection method based on a modified YOLOv4,...To tackle the problem of aquatic environment pollution,a vision-based autonomous underwater garbage cleaning robot has been developed in our laboratory.We propose a garbage detection method based on a modified YOLOv4,allowing high-speed and high-precision object detection.Specifically,the YOLOv4 algorithm is chosen as a basic neural network framework to perform object detection.With the purpose of further improvement on the detection accuracy,YOLOv4 is transformed into a four-scale detection method.To improve the detection speed,model pruning is applied to the new model.By virtue of the improved detection methods,the robot can collect garbage autonomously.The detection speed is up to 66.67 frames/s with a mean average precision(mAP)of 95.099%,and experimental results demonstrate that both the detection speed and the accuracy of the improved YOLOv4 are excellent.展开更多
For complex functions to emerge in artificial systems,it is important to understand the intrinsic mechanisms of biological swarm behaviors in nature.In this paper,we present a comprehensive survey of pursuit–evasion,...For complex functions to emerge in artificial systems,it is important to understand the intrinsic mechanisms of biological swarm behaviors in nature.In this paper,we present a comprehensive survey of pursuit–evasion,which is a critical problem in biological groups.First,we review the problem of pursuit–evasion from three different perspectives:game theory,control theory and artificial intelligence,and bio-inspired perspectives.Then we provide an overview of the research on pursuit–evasion problems in biological systems and artificial systems.We summarize predator pursuit behavior and prey evasion behavior as predator–prey behavior.Next,we analyze the application of pursuit–evasion in artificial systems from three perspectives,i.e.,strong pursuer group vs.weak evader group,weak pursuer group vs.strong evader group,and equal-ability group.Finally,relevant prospects for future pursuit–evasion challenges are discussed.This survey provides new insights into the design of multi-agent and multi-robot systems to complete complex hunting tasks in uncertain dynamic scenarios.展开更多
This paper investigates two distributed accelerated primal-dual neurodynamic approaches over undirected connected graphs for resource allocation problems(RAP)where the objective functions are generally convex.With the...This paper investigates two distributed accelerated primal-dual neurodynamic approaches over undirected connected graphs for resource allocation problems(RAP)where the objective functions are generally convex.With the help of projection operators,a primal-dual framework,and Nesterov's accelerated method,we first design a distributed accelerated primal-dual projection neurodynamic approach(DAPDP),and its convergence rate of the primal-dual gap is O(1/(t^(2)))by selecting appropriate parameters and initial values.Then,when the local closed convex sets are convex inequalities which have no closed-form solutions of their projection operators,we further propose a distributed accelerated penalty primal-dual neurodynamic approach(DAPPD)on the strength of the penalty method,primal-dual framework,and Nesterov's accelerated method.Based on the above analysis,we prove that DAPPD also has a convergence rate O(1/(t^(2)))of the primal-dual gap.Compared with the distributed dynamical approaches based on the classical primal-dual framework,our proposed distributed accelerated neurodynamic approaches have faster convergence rates.Numerical simulations demonstrate that our proposed neurodynamic approaches are feasible and effective.展开更多
The game of Tibetan Go faces the scarcity of expert knowledge and research literature.Therefore,we study the zero learning model of Tibetan Go under limited computing power resources and propose a novel scaleinvariant...The game of Tibetan Go faces the scarcity of expert knowledge and research literature.Therefore,we study the zero learning model of Tibetan Go under limited computing power resources and propose a novel scaleinvariant U-Net style two-headed output lightweight network TibetanGoTinyNet.The lightweight convolutional neural networks and capsule structure are applied to the encoder and decoder of TibetanGoTinyNet to reduce computational burden and achieve better feature extraction results.Several autonomous self-attention mechanisms are integrated into TibetanGoTinyNet to capture the Tibetan Go board’s spatial and global information and select important channels.The training data are generated entirely from self-play games.TibetanGoTinyNet achieves 62%–78%winning rate against other four U-Net style models including Res-UNet,Res-UNet Attention,Ghost-UNet,and Ghost Capsule-UNet.It also achieves 75%winning rate in the ablation experiments on the attention mechanism with embedded positional information.The model saves about 33%of the training time with 45%–50%winning rate for different Monte–Carlo tree search(MCTS)simulation counts when migrated from 9×9 to 11×11 boards.Code for our model is available at https://github.com/paulzyy/TibetanGoTinyNet.展开更多
Underwater robotic operation usually requires visual perception(e.g.,object detection and tracking),but underwater scenes have poor visual quality and represent a special domain which can affect the accuracy of visual...Underwater robotic operation usually requires visual perception(e.g.,object detection and tracking),but underwater scenes have poor visual quality and represent a special domain which can affect the accuracy of visual perception.In addition,detection continuity and stability are important for robotic perception,but the commonly used static accuracy based evaluation(i.e.,average precision)is insufficient to reflect detector performance across time.In response to these two problems,we present a design for a novel robotic visual perception framework.First,we generally investigate the relationship between a quality-diverse data domain and visual restoration in detection performance.As a result,although domain quality has an ignorable effect on within-domain detection accuracy,visual restoration is beneficial to detection in real sea scenarios by reducing the domain shift.Moreover,non-reference assessments are proposed for detection continuity and stability based on object tracklets.Further,online tracklet refinement is developed to improve the temporal performance of detectors.Finally,combined with visual restoration,an accurate and stable underwater robotic visual perception framework is established.Small-overlap suppression is proposed to extend video object detection(VID)methods to a single-object tracking task,leading to the flexibility to switch between detection and tracking.Extensive experiments were conducted on the ImageNet VID dataset and real-world robotic tasks to verify the correctness of our analysis and the superiority of our proposed approaches.The codes are available at https://github.com/yrqs/VisPerception.展开更多
The existing fixed gait lower limb rehabilitation robots perform a predetermined walking trajectory for patients,ignoring their residual muscle strength.To enhance patient participation and safety in training,this pap...The existing fixed gait lower limb rehabilitation robots perform a predetermined walking trajectory for patients,ignoring their residual muscle strength.To enhance patient participation and safety in training,this paper aims to develop a lower limb rehabilitation robot with adaptive gait training capability relying on human–robot interaction force measurement.Firstly,a novel lower limb rehabilitation robot system with several active and passive driven joints is developed,and 2 face-to-face mounted cantilever beam force sensors are employed to measure the human–robot interaction forces.Secondly,a dynamic model of the rehabilitation training robot is constructed to estimate the driven forces of the human lower leg in a completely passive state.Thereafter,based on the theoretical moment from the dynamics and the actual joint interaction force collected by the sensors,an adaptive gait adjustment method is proposed to achieve the goal of adapting to the wearer’s movement intention.Finally,interactive experiments are carried out to validate the effectiveness of the developed rehabilitation training robot system.The proposed rehabilitation training robot system with adaptive gaits offers great potential for future highquality rehabilitation training,e.g.,improving participation and safety.展开更多
文摘Pedestrian trajectory prediction can significantly enhance the perception and decision-making capabilities of autonomous driving systems and intelligent surveillance systems based on camera sensors by predicting the states and behavior intentions of surrounding pedestrians.However,existing trajectory prediction methods remain failing to effectively model the diverse and complex interactions in the real world,including pedestrian-pedestrian interactions and pedestrian-environment interactions.Besides,these methods are not effective in capturing and characterizing the multimodal property of future trajectories.To address these challenges above,we propose to devise a handdesigned graph convolution and spatial cross attention to dynamically capture the diverse spatial interactions between pedestrians.To effectively explore the impact of scenarios on pedestrian trajectory,we build a pedestrian map,which can reflect the scene constraints and pedestrian motion preferences.Meanwhile,we construct a trajectory multimodality-aware module to capture the different potential mode implicit in diverse social behaviors for pedestrian future trajectory uncertainty.Finally,we compared the proposed method with trajectory prediction baselines on commonly used public pedestrian benchmarks,demonstrating the superior performance of our approach.
基金supported by the National Natural Science Foundation of China(62302047,62203250)the Science and Technology Development Fund of Macao SAR(0093/2023/RIA2,0050/2020/A1).
文摘THE tremendous impact of large models represented by ChatGPT[1]-[3]makes it necessary to con-sider the practical applications of such models[4].However,for an artificial intelligence(AI)to truly evolve,it needs to possess a physical“body”to transition from the virtual world to the real world and evolve through interaction with the real environments.In this context,“embodied intelligence”has sparked a new wave of research and technology,leading AI beyond the digital realm into a new paradigm that can actively act and perceive in a physical environment through tangible entities such as robots and automated devices[5].
文摘Origami and kirigami structures originate from the ancient art of paper folding and cutting.They offer the opportunity to achieve superior performances and unusual function-alities brought by complex three-dimensional geometries using approachable manufacturing technologies that work on two-dimensional sheets.
基金This work was supported in part by the National Natural Science Foundation of China(U1909206,61725305,61903007,62073196)in part by the S&T Program of Hebei(F2020203037).
文摘As a cross-cutting field between ocean development and multi-robot system(MRS),the underwater multi-robot system(UMRS)has gained increasing attention from researchers and engineers in recent decades.In this paper,we present a comprehensive survey of cooperation issues,one of the key components of UMRS,from the perspective of the emergence of new functions.More specifically,we categorize the cooperation in terms of task-space,motion-space,measurement-space,as well as their combination.Further,we analyze the architecture of UMRS from three aspects,i.e.,the performance of the individual underwater robot,the new functions of underwater robots,and the technical approaches of MRS.To conclude,we have discussed related promising directions for future research.This survey provides valuable insight into the reasonable utilization of UMRS to attain diverse underwater tasks in complex ocean application scenarios.
基金supported in part by the National Natural Science Foundation of China(61873304,62173048,62106023)the Key Science and Technology Projects of Jilin Province,China(20210201106GX)+2 种基金the Innovation and Entrepreneurship Talent funding Project of Jilin Province(2022QN04)the Changchun Science and Technology Project(21ZY41)Beijing Natural Science Foundation(2022MQ05)。
文摘Dear Editor,Quadratic programming problems(QPs)receive a lot of attention in various fields of science computing and engineering applications,such as manipulator control[1].Recursive neural network(RNN)is considered to be a powerful QPs solver due to its parallel processing capability and feasibility of hardware implementation[2].
基金National Key Research and Development Program of China(2020YFE0204200,2022YFB4701900)National Natural Science Foundation of China(11988102,12202008)Experiments for Space Exploration Program and the Qian Xuesen Laboratory,China Academy of Space Technology(TKTSPY-2020-03-05).
文摘Instability-induced wrinkle patterns of thin sheets are ubiquitous in nature,which often result in origami-like patterns that provide inspiration for the engineering of origami designs.Inspired by instability-induced origami patterns,we propose a computational origami design method based on the nonlinear analysis of loaded thin sheets and topology optimization.The bar-and-hinge model is employed for the nonlinear structural analysis,added with a displacement perturbation strategy to initiate out-of-plane buckling.Borrowing ideas from topology optimization,a continuous crease indicator is introduced as the design variable to indicate the state of a crease,which is penalized by power functions to establish the mapping relationships between the crease indicator and hinge properties.Minimizing the structural strain energy with a crease length constraint,we are able to evolve a thin sheet into an origami structure with an optimized crease pattern.Two examples with different initial setups are illustrated,demonstrating the effectiveness and feasibility of the method.
基金supported by the National Natural Science Foundation of China(No.12372259).
文摘Polyelectrolyte(PE)gels,distinguished by their unique stimuli-responsive swelling behavior,serve as the basis of broad applications,such as artificial muscles and drug delivery.In this work,we present a theoretical model to analyze the electrostatics and its contribution to the swelling behavior of PE gels in salt solutions.By minimizing the free energy of PE gels,we obtain two distinct scaling regimes for the swelling ratio at equilibrium with respect to the salt concentration.We compare our predictions for the swelling ratio with experimental measurements,which show excellent agreement.In addition,we employ a finite element method to assess the applicability range of our theoretical model and assumptions.We anticipate that our model will also provide valuable insights into drug adsorption and release,deformation of red blood cells,4D printing and soft robotics,where the underlying mechanism of swelling remains enigmatic.
基金supported in part by the National Natural Science Foundation of China (62373065,61873304,62173048,62106023)the Innovation and Entrepreneurship Talent funding Project of Jilin Province(2022QN04)+1 种基金the Changchun Science and Technology Project (21ZY41)the Open Research Fund of National Mobile Communications Research Laboratory,Southeast University (2024D09)。
文摘This paper presents a distributed scheme with limited communications, aiming to achieve cooperative motion control for multiple omnidirectional mobile manipulators(MOMMs).The proposed scheme extends the existing single-agent motion control to cater to scenarios involving the cooperative operation of MOMMs. Specifically, squeeze-free cooperative load transportation is achieved for the end-effectors of MOMMs by incorporating cooperative repetitive motion planning(CRMP), while guiding each individual to desired poses. Then, the distributed scheme is formulated as a time-varying quadratic programming(QP) and solved online utilizing a noise-tolerant zeroing neural network(NTZNN). Theoretical analysis shows that the NTZNN model converges globally to the optimal solution of QP in the presence of noise. Finally, the effectiveness of the control design is demonstrated by numerical simulations and physical platform experiments.
基金supported by the National Key Research and Development Program of China(Grant No.2022YFA1203100)the National Natural Science Foundation of China(Grant No.52076002)+1 种基金the High-performance Computing Platform of Peking Universitysupport from the New Cornerstone Science Foundation through the XPLORER PRIZE。
文摘Disordered hyperuniformity(DHU)is a recently discovered novel state of amorphous systems characterized by strongly suppressed density fluctuations at large length scales as in crystalline materials,which offers great potential for achieving unusual mechanical,electronic,and photonic properties.However,despite the fundamental and technological importance of thermal transport in amorphous solids,the effect of DHU remains largely unexplored.Here,we theoretically study thermal transport in a class of two-dimensional DHU materials—monolayer amorphous carbon(MAC).Beginning with a perfect graphene lattice,we continuously apply Stone-Wales transformations to generate a series of MAC models with varied degrees of disorder and defects,which are quantified through comprehensive structural analysis including the so-called hyperuniformity index(H),where a smaller H indicates a higher degree of hyperuniformity.Subsequently,we conduct molecular dynamics simulations to obtain the thermal conductivity(κ).A significant correlation between the thermal transport behavior and degree of hyperuniformity is observed,with the room-temperatureκdecreasing from 26.3 to 5.3 W m^(-1)K^(-1)while H is tuned from 0.0004 to 0.024.Remarkably,two distinct transport regimes are identified,including a nearly-DHU regime at small H(<0.005)whereκdrops sharply and a non-DHU region at larger H whereκbecomes relatively flat.Mode-resolved analysis reveals longer lifetime and higher participation ratio for the heat carriers in nearly-DHU MAC,implying that the hidden long-range correlations could support extended modes that enhance transport.Our work highlights the impact of DHU on the thermal properties of amorphous materials and represents a conceptual advancement that is worthy of future exploration.
基金supported in part by Beijing Natural Science Foundation(Nos.L233030 and 2022MQ05)in part by the National Natural Science Foundation of China(Nos.62073322,61836015,and 61633020).
文摘Few-shot object detection receives much attention with the ability to detect novel class objects using limited annotated data.The transfer learning-based solution becomes popular due to its simple training with good accuracy,however,it is still challenging to enrich the feature diversity during the training process.And fine-grained features are also insufficient for novel class detection.To deal with the problems,this paper proposes a novel few-shot object detection method based on dual-domain feature fusion and patch-level attention.Upon original base domain,an elementary domain with more category-agnostic features is superposed to construct a two-stream backbone,which benefits to enrich the feature diversity.To better integrate various features,a dual-domain feature fusion is designed,where the feature pairs with the same size are complementarily fused to extract more discriminative features.Moreover,a patch-wise feature refinement termed as patch-level attention is presented to mine internal relations among the patches,which enhances the adaptability to novel classes.In addition,a weighted classification loss is given to assist the fine-tuning of the classifier by combining extra features from FPN of the base training model.In this way,the few-shot detection quality to novel class objects is improved.Experiments on PASCAL VOC and MS COCO datasets verify the effectiveness of the method.
基金supported in part by the National Natural Science Foundation of China(T2121002,U1909206,61903007,62073196)and in part by the S&T Program of Hebei(F2020203037).
文摘The task of path planning in amphibious environments requires additional consideration due to the complexity of the amphibious environments.This paper presents a path planning method for an amphibious robot named\AmphiRobot"with its dynamic constraints considered.First,an explicit dynamic model using Kane's method is presented.The hydrodynamic parameters are obtained through computational°uid dynamics simulations.Furthermore,a path planning method based on a hybrid¯reworks algorithm is proposed,combining the¯reworks algorithm and bare bones¯reworks algorithm,aiming at the amphibious robot's characteristics of multiple motion modes and working environments.The initially planned path is then smoothed using Dubins path under constraints determined by the dynamic model.Simulation reveals that the performance of the hybrid¯reworks algorithm approach is better than the¯reworks algorithm and bare bones¯reworks algorithm is applied separately in the amphibious environment scenarios.
基金supported by the National Natural Science Foundation of China(61836015,U1909206,62022090,and 62033013).
文摘Central pattern generators(CPGs)have been widely applied in robot motion control for the spontaneous output of coherent periodic rhythms.However,the underlying CPG network exhibits good convergence performance only within a certain range of parameter spaces,and the coupling of oscillators affects the network output accuracy in complex topological relationships.Moreover,CPGs may diverge when parameters change drastically,and the divergence is irreversible,which is catastrophic for the control of robot motion.Therefore,normalized asymmetric CPGs(NA-CPGs)that normalize the amplitude parameters of Hopf-based CPGs and add a constraint function and a frequency regulation mechanism are proposed.NA-CPGs can realize parameter decoupling,precise amplitude output,and stable and rapid convergence,as well as asymmetric output waveforms.Thus,it can effectively cope with large parameter changes to avoid network oscillations and divergence.To optimize the parameters of the NA-CPG model,a reinforcement-learning-based online optimization method is further proposed.Meanwhile,a biomimetic robotic fish is illustrated to realize the whole optimization process.Simulations demonstrated that the designed NA-CPGs exhibit stable,secure,and accurate network outputs,and the proposed optimization method effectively improves the swimming speed and reduces the lateral swing of the multijoint robotic fish by 6.7%and 41.7%,respectively.The proposed approach provides a significant improvement in CPG research and can be widely employed in the field of robot motion control.
基金National Natural Science Foundation of China,Grant/Award Numbers:11972185,12032010。
文摘To address the challenging task of effective sound absorption in the low and broad frequency band for underwater structures,we propose a novel grating‐like anechoic layer by filling rubber blocks and an air backing layer into metallic grating.The metallic gratings are incorporated into the anechoic layer as a skeleton for enhanced viscoelastic dissipation by promoting shear deformation between rubber and metal plates.The introduction of an air backing layer releases the bottom constraint of the rubber,thus intensifying its deformation under acoustic excitation.Based on the homogenization method and the transfer matrix method,a theoretical model is developed to evaluate the sound absorption performance of the proposed anechoic layer,which is validated against finite element simulation results.It is demonstrated that a sound absorption coefficient of the grating‐like anechoic layer of 0.8 can be achieved in the frequency range of 1294-10000 Hz.Given the importance of sound absorption at varying frequencies,the weighted average method is subsequently used to comprehensively evaluate the performance of the anechoic layer.Then,with structural density taken into consideration,an integrated index is proposed to further evaluate the acoustic properties of the proposed anechoic layer.Finally,the backing conditions and the boundary conditions of finite‐size structures are discussed.The results provide helpful theoretical guidance for designing novel acoustic metamaterials with broadband low‐frequency underwater sound absorption.
基金This work was supported by the National Natural Science Foundation of China(Nos.62073322 and 61633020)the CIE-Tencent Robotics X Rhino-Bird Focused Research Program(No.2022-07)the Beijing Natural Science Foundation(No.2022MQ05).
文摘Grasp detection plays a critical role for robot manipulation.Mainstream pixel-wise grasp detection networks with encoder-decoder structure receive much attention due to good accuracy and efficiency.However,they usually transmit the high-level feature in the encoder to the decoder,and low-level features are neglected.It is noted that low-level features contain abundant detail information,and how to fully exploit low-level features remains unsolved.Meanwhile,the channel information in high-level feature is also not well mined.Inevitably,the performance of grasp detection is degraded.To solve these problems,we propose a grasp detection network with hierarchical multi-scale feature fusion and inverted shuffle residual.Both low-level and high-level features in the encoder are firstly fused by the designed skip connections with attention module,and the fused information is then propagated to corresponding layers of the decoder for in-depth feature fusion.Such a hierarchical fusion guarantees the quality of grasp prediction.Furthermore,an inverted shuffle residual module is created,where the high-level feature from encoder is split in channel and the resultant split features are processed in their respective branches.By such differentiation processing,more high-dimensional channel information is kept,which enhances the representation ability of the network.Besides,an information enhancement module is added before the encoder to reinforce input information.The proposed method attains 98.9%and 97.8%in image-wise and object-wise accuracy on the Cornell grasping dataset,respectively,and the experimental results verify the effectiveness of the method.
基金supported by the National Natural Science Foundation of China(Nos.61725305,U1909206,T2121002,and62073196)the Postdoctoral Innovative Talent Support Program(No.BX2021010)the S&T Program of Hebei Province,China(No.F2020203037)。
文摘To tackle the problem of aquatic environment pollution,a vision-based autonomous underwater garbage cleaning robot has been developed in our laboratory.We propose a garbage detection method based on a modified YOLOv4,allowing high-speed and high-precision object detection.Specifically,the YOLOv4 algorithm is chosen as a basic neural network framework to perform object detection.With the purpose of further improvement on the detection accuracy,YOLOv4 is transformed into a four-scale detection method.To improve the detection speed,model pruning is applied to the new model.By virtue of the improved detection methods,the robot can collect garbage autonomously.The detection speed is up to 66.67 frames/s with a mean average precision(mAP)of 95.099%,and experimental results demonstrate that both the detection speed and the accuracy of the improved YOLOv4 are excellent.
基金Project supported by the National Natural Science Foundation of China(Nos.U1909206,T2121002,61903007,and 11972373)。
文摘For complex functions to emerge in artificial systems,it is important to understand the intrinsic mechanisms of biological swarm behaviors in nature.In this paper,we present a comprehensive survey of pursuit–evasion,which is a critical problem in biological groups.First,we review the problem of pursuit–evasion from three different perspectives:game theory,control theory and artificial intelligence,and bio-inspired perspectives.Then we provide an overview of the research on pursuit–evasion problems in biological systems and artificial systems.We summarize predator pursuit behavior and prey evasion behavior as predator–prey behavior.Next,we analyze the application of pursuit–evasion in artificial systems from three perspectives,i.e.,strong pursuer group vs.weak evader group,weak pursuer group vs.strong evader group,and equal-ability group.Finally,relevant prospects for future pursuit–evasion challenges are discussed.This survey provides new insights into the design of multi-agent and multi-robot systems to complete complex hunting tasks in uncertain dynamic scenarios.
基金supported by the National Natural Science Foundation of China (Grant No.62176218)the Fundamental Research Funds for the Central Universities (Grant No.XDJK2020TY003)。
文摘This paper investigates two distributed accelerated primal-dual neurodynamic approaches over undirected connected graphs for resource allocation problems(RAP)where the objective functions are generally convex.With the help of projection operators,a primal-dual framework,and Nesterov's accelerated method,we first design a distributed accelerated primal-dual projection neurodynamic approach(DAPDP),and its convergence rate of the primal-dual gap is O(1/(t^(2)))by selecting appropriate parameters and initial values.Then,when the local closed convex sets are convex inequalities which have no closed-form solutions of their projection operators,we further propose a distributed accelerated penalty primal-dual neurodynamic approach(DAPPD)on the strength of the penalty method,primal-dual framework,and Nesterov's accelerated method.Based on the above analysis,we prove that DAPPD also has a convergence rate O(1/(t^(2)))of the primal-dual gap.Compared with the distributed dynamical approaches based on the classical primal-dual framework,our proposed distributed accelerated neurodynamic approaches have faster convergence rates.Numerical simulations demonstrate that our proposed neurodynamic approaches are feasible and effective.
基金the National Natural Science Foundation of China(Nos.62276285 and 62236011)the Major Projects of Social Science Fundation of China(No.20&ZD279)。
文摘The game of Tibetan Go faces the scarcity of expert knowledge and research literature.Therefore,we study the zero learning model of Tibetan Go under limited computing power resources and propose a novel scaleinvariant U-Net style two-headed output lightweight network TibetanGoTinyNet.The lightweight convolutional neural networks and capsule structure are applied to the encoder and decoder of TibetanGoTinyNet to reduce computational burden and achieve better feature extraction results.Several autonomous self-attention mechanisms are integrated into TibetanGoTinyNet to capture the Tibetan Go board’s spatial and global information and select important channels.The training data are generated entirely from self-play games.TibetanGoTinyNet achieves 62%–78%winning rate against other four U-Net style models including Res-UNet,Res-UNet Attention,Ghost-UNet,and Ghost Capsule-UNet.It also achieves 75%winning rate in the ablation experiments on the attention mechanism with embedded positional information.The model saves about 33%of the training time with 45%–50%winning rate for different Monte–Carlo tree search(MCTS)simulation counts when migrated from 9×9 to 11×11 boards.Code for our model is available at https://github.com/paulzyy/TibetanGoTinyNet.
基金Project supported by the National Natural Science Foundation of China(Nos.61633004,61725305,and 62073196)the S&T Program of Hebei Province,China(No.F2020203037)。
文摘Underwater robotic operation usually requires visual perception(e.g.,object detection and tracking),but underwater scenes have poor visual quality and represent a special domain which can affect the accuracy of visual perception.In addition,detection continuity and stability are important for robotic perception,but the commonly used static accuracy based evaluation(i.e.,average precision)is insufficient to reflect detector performance across time.In response to these two problems,we present a design for a novel robotic visual perception framework.First,we generally investigate the relationship between a quality-diverse data domain and visual restoration in detection performance.As a result,although domain quality has an ignorable effect on within-domain detection accuracy,visual restoration is beneficial to detection in real sea scenarios by reducing the domain shift.Moreover,non-reference assessments are proposed for detection continuity and stability based on object tracklets.Further,online tracklet refinement is developed to improve the temporal performance of detectors.Finally,combined with visual restoration,an accurate and stable underwater robotic visual perception framework is established.Small-overlap suppression is proposed to extend video object detection(VID)methods to a single-object tracking task,leading to the flexibility to switch between detection and tracking.Extensive experiments were conducted on the ImageNet VID dataset and real-world robotic tasks to verify the correctness of our analysis and the superiority of our proposed approaches.The codes are available at https://github.com/yrqs/VisPerception.
基金supported in part by the National Natural Science Foundation of China under Grants 62373353,and 62033013in part by Youth Innovation Promotion Association CAS(2019138).
文摘The existing fixed gait lower limb rehabilitation robots perform a predetermined walking trajectory for patients,ignoring their residual muscle strength.To enhance patient participation and safety in training,this paper aims to develop a lower limb rehabilitation robot with adaptive gait training capability relying on human–robot interaction force measurement.Firstly,a novel lower limb rehabilitation robot system with several active and passive driven joints is developed,and 2 face-to-face mounted cantilever beam force sensors are employed to measure the human–robot interaction forces.Secondly,a dynamic model of the rehabilitation training robot is constructed to estimate the driven forces of the human lower leg in a completely passive state.Thereafter,based on the theoretical moment from the dynamics and the actual joint interaction force collected by the sensors,an adaptive gait adjustment method is proposed to achieve the goal of adapting to the wearer’s movement intention.Finally,interactive experiments are carried out to validate the effectiveness of the developed rehabilitation training robot system.The proposed rehabilitation training robot system with adaptive gaits offers great potential for future highquality rehabilitation training,e.g.,improving participation and safety.
