Robot learning in unstructured environments has been proved to be an extremely challenging problem, mainly because of many uncertainties always present in the real world. Human beings, on the other hand, seem to cope ...Robot learning in unstructured environments has been proved to be an extremely challenging problem, mainly because of many uncertainties always present in the real world. Human beings, on the other hand, seem to cope very well with uncertain and unpredictable environments, often relying on perception-based information. Furthermore, humans beings can also utilize perceptions to guide their learning on those parts of the perception-action space that are actually relevant to the task. Therefore, we conduct a research aimed at improving robot learning through the incorporation of both perception-based and measurement-based information. For this reason, a fuzzy reinforcement learning (FRL) agent is proposed in this paper. Based on a neural-fuzzy architecture, different kinds of information can be incorporated into the FRL agent to initialise its action network, critic network and evaluation feedback module so as to accelerate its learning. By making use of the global optimisation capability of GAs (genetic algorithms), a GA-based FRL (GAFRL) agent is presented to solve the local minima problem in traditional actor-critic reinforcement learning. On the other hand, with the prediction capability of the critic network, GAs can perform a more effective global search. Different GAFRL agents are constructed and verified by using the simulation model of a physical biped robot. The simulation analysis shows that the biped learning rate for dynamic balance can be improved by incorporating perception-based information on biped balancing and walking evaluation. The biped robot can find its application in ocean exploration, detection or sea rescue activity, as well as military maritime activity.展开更多
Humans achieve cognitive development through continuous interaction with their environment,enhancing both perception and behavior.However,current robots lack the capacity for human-like action and evolution,posing a b...Humans achieve cognitive development through continuous interaction with their environment,enhancing both perception and behavior.However,current robots lack the capacity for human-like action and evolution,posing a bottleneck to improving robotic intelligence.Existing research predominantly models robots as one-way,static mappings from observations to actions,neglecting the dynamic processes of perception and behavior.This paper introduces a novel approach to robot cognitive learning by considering physical properties.We propose a theoretical framework wherein a robot is conceptualized as a three-body physical system comprising a perception-body(P-body),a cognition-body(C-body),and a behavior-body(B-body).Each body engages in physical dynamics and operates within a closed-loop interaction.Significantly,three crucial interactions connect these bodies.The C-body relies on the Pbody's extracted states and reciprocally offers long-term rewards,optimizing the P-body's perception policy.In addition,the C-body directs the B-body's actions through sub-goals,and subsequent P-body-derived states facilitate the C-body's cognition dynamics learning.At last,the B-body would follow the sub-goal generated by the C-body and perform actions conditioned on the perceptive state from the P-body,which leads to the next interactive step.These interactions foster the joint evolution of each body,culminating in optimal design.To validate our approach,we employ a navigation task using a four-legged robot,D'Kitty,equipped with a movable global camera.Navigational prowess demands intricate coordination of sensing,planning,and D'Kitty's motion.Leveraging our framework yields superior task performance compared with conventional methodologies.In conclusion,this paper establishes a paradigm shift in robot cognitive learning by integrating physical interactions across the P-body,C-body,and B-body,while considering physical properties.Our framework's successful application to a navigation task underscores its efficacy in enhancing robotic intelligence.展开更多
The skill of robotic hand-eye coordination not only helps robots to deal with real time environment,but also afects the fundamental framework of robotic cognition.A number of approaches have been developed in the lite...The skill of robotic hand-eye coordination not only helps robots to deal with real time environment,but also afects the fundamental framework of robotic cognition.A number of approaches have been developed in the literature for construction of the robotic hand-eye coordination.However,several important features within infant developmental procedure have not been introduced into such approaches.This paper proposes a new method for robotic hand-eye coordination by imitating the developmental progress of human infants.The work employs a brain-like neural network system inspired by infant brain structure to learn hand-eye coordination,and adopts a developmental mechanism from psychology to drive the robot.The entire learning procedure is driven by developmental constraint： The robot starts to act under fully constrained conditions,when the robot learning system becomes stable,a new constraint is assigned to the robot.After that,the robot needs to act with this new condition again.When all the contained conditions have been overcome,the robot is able to obtain hand-eye coordination ability.The work is supported by experimental evaluation,which shows that the new approach is able to drive the robot to learn autonomously,and make the robot also exhibit developmental progress similar to human infants.展开更多
Collaborative heterogeneous robot systems can greatly enhance the efficiency of target search and navigation tasks.In this paper,we design a heterogeneous robot system consisting of an unmanned aerial vehicle(UAV)and ...Collaborative heterogeneous robot systems can greatly enhance the efficiency of target search and navigation tasks.In this paper,we design a heterogeneous robot system consisting of an unmanned aerial vehicle(UAV)and an unmanned ground vehicle(UGV)for search and rescue missions in unknown environments.The system is able to search for targets and navigate to them in a maze-like mine environment with the policies learned through deep reinforcement learning algorithms.During the training process,if two robots are trained simultaneously,the rewards related to their collaboration may not be properly obtained.Hence,we introduce a multi-stage reinforcement learning framework and a curiosity module to encourage agents to explore unvisited environments.Experiments in simulation environments show that our framework can train the heterogeneous robot system to achieve the search and navigation with unknown target locations while existing baselines may not.The UGV achieves a success rate of 89.1%in the mission within the original environment,and maintains a 67.6%success rate in untrained complex environments.展开更多
Robots are key to expanding the scope of space applications.The end-to-end training for robot vision-based detection and precision operations is challenging owing to constraints such as extreme environments and high c...Robots are key to expanding the scope of space applications.The end-to-end training for robot vision-based detection and precision operations is challenging owing to constraints such as extreme environments and high computational overhead.This study proposes a lightweight integrated framework for grasp detection and imitation learning,named GD-IL;it comprises a grasp detection algorithm based on manipulability and Gaussian mixture model(manipulability-GMM),and a grasp trajectory generation algorithm based on a two-stage robot imitation learning algorithm(TS-RIL).In the manipulability-GMM algorithm,we apply GMM clustering and ellipse regression to the object point cloud,propose two judgment criteria to generate multiple candidate grasp bounding boxes for the robot,and use manipulability as a metric for selecting the optimal grasp bounding box.The stages of the TS-RIL algorithm are grasp trajectory learning and robot pose optimization.In the first stage,the robot grasp trajectory is characterized using a second-order dynamic movement primitive model and Gaussian mixture regression(GMM).By adjusting the function form of the forcing term,the robot closely approximates the target-grasping trajectory.In the second stage,a robot pose optimization model is built based on the derived pose error formula and manipulability metric.This model allows the robot to adjust its configuration in real time while grasping,thereby effectively avoiding singularities.Finally,an algorithm verification platform is developed based on a Robot Operating System and a series of comparative experiments are conducted in real-world scenarios.The experimental results demonstrate that GD-IL significantly improves the effectiveness and robustness of grasp detection and trajectory imitation learning,outperforming existing state-of-the-art methods in execution efficiency,manipulability,and success rate.展开更多
Endowing quadruped robots with the skill to forward jump is conducive to making it overcome barriers and pass through complex terrains.In this paper,a model-free control architecture with target-guided policy optimiza...Endowing quadruped robots with the skill to forward jump is conducive to making it overcome barriers and pass through complex terrains.In this paper,a model-free control architecture with target-guided policy optimization and deep reinforcement learn-ing(DRL)for quadruped robot jumping is presented.First,the jumping phase is divided into take-off and flight-landing phases,and op-timal strategies with soft actor-critic(SAC)are constructed for the two phases respectively.Second,policy learning including expecta-tions,penalties in the overall jumping process,and extrinsic excitations is designed.Corresponding policies and constraints are all provided for successful take-off,excellent flight attitude and stable standing after landing.In order to avoid low efficiency of random ex-ploration,a curiosity module is introduced as extrinsic rewards to solve this problem.Additionally,the target-guided module encour-ages the robot explore closer and closer to desired jumping target.Simulation results indicate that the quadruped robot can realize com-pleted forward jumping locomotion with good horizontal and vertical distances,as well as excellent motion attitudes.展开更多
Bagging is an essential skill that humans perform in their daily activities.However,deformable objects,such as bags,are complex for robots to manipulate.A learning-based framework that enables robots to learn bagging ...Bagging is an essential skill that humans perform in their daily activities.However,deformable objects,such as bags,are complex for robots to manipulate.A learning-based framework that enables robots to learn bagging is presented.The novelty of this framework is its ability to learn and perform bagging without relying on simulations.The learning process is accomplished through a reinforcement learning(RL)algorithm introduced and designed to find the best grasping points of the bag based on a set of compact state representations.The framework utilises a set of primitive actions and represents the task in five states.In our experiments,the framework reached 60% and 80% success rates after around 3 h of training in the real world when starting the bagging task from folded and unfolded states,respectively.Finally,the authors test the trained RL model with eight more bags of different sizes to evaluate its generalisability.展开更多
Reactive planning and control capacity for collaborative robots is essential when the tasks change online in an unstructured environment.This is more difficult for collaborative mobile manipulators(CMM)due to high red...Reactive planning and control capacity for collaborative robots is essential when the tasks change online in an unstructured environment.This is more difficult for collaborative mobile manipulators(CMM)due to high redundancies.To this end,this paper proposed a reactive whole-body locomotion-integrated manipulation approach based on combined learning and optimization.First,human demonstrations are collected,where the wrist and pelvis movements are treated as whole-body trajectories,mapping to the end-effector(EE)and the mobile base(MB)of CMM,respectively.A time-input kernelized movement primitive(T-KMP)learns the whole-body trajectory,and a multi-dimensional kernelized movement primitive(M-KMP)learns the spatial relationship between the MB and EE pose.According to task changes,the T-KMP adapts the learned trajectories online by inserting the new desired point predicted by MKMP.Then,the updated reference trajectories are sent to a hierarchical quadratic programming(HQP)controller,where the EE and the MB trajectories tracking are set as the first and second priority tasks,generating the feasible and optimal joint level commands.An ablation simulation experiment with CMM of the HQP is conducted to show the necessity of MB trajectory tracking in mimicking human whole-body motion behavior.Finally,the tasks of the reactive pick-and-place and reactive reaching were undertaken,where the target object was randomly moved,even out of the region of demonstrations.The results showed that the proposed approach can successfully transfer and adapt the human whole-body loco-manipulation skills to CMM online with task changes.展开更多
In this article,a robot skills learning framework is developed,which considers both motion modeling and execution.In order to enable the robot to learn skills from demonstrations,a learning method called dynamic movem...In this article,a robot skills learning framework is developed,which considers both motion modeling and execution.In order to enable the robot to learn skills from demonstrations,a learning method called dynamic movement primitives(DMPs)is introduced to model motion.A staged teaching strategy is integrated into DMPs frameworks to enhance the generality such that the complicated tasks can be also performed for multi-joint manipulators.The DMP connection method is used to make an accurate and smooth transition in position and velocity space to connect complex motion sequences.In addition,motions are categorized into different goals and durations.It is worth mentioning that an adaptive neural networks(NNs)control method is proposed to achieve highly accurate trajectory tracking and to ensure the performance of action execution,which is beneficial to the improvement of reliability of the skills learning system.The experiment test on the Baxter robot verifies the effectiveness of the proposed method.展开更多
Real-time proprioception presents a significant challenge for soft robots due to their infinite degrees of freedom and intrinsic compliance.Previous studies mostly focused on specific sensors and actuators.There is st...Real-time proprioception presents a significant challenge for soft robots due to their infinite degrees of freedom and intrinsic compliance.Previous studies mostly focused on specific sensors and actuators.There is still a lack of generalizable technologies for integrating soft sensing elements into soft actuators and mapping sensor signals to proprioception parameters.To tackle this problem,we employed multi-material 3D printing technology to fabricate sensorized soft-bending actuators(SBAs)using plain and conductive thermoplastic polyurethane(TPU)filaments.We designed various geometric shapes for the sensors and investigated their strain-resistive performance during deformation.To address the nonlinear time-variant behavior of the sensors during dynamic modeling,we adopted a data-driven approach using different deep neural networks to learn the relationship between sensor signals and system states.A series of experiments in various actuation scenarios were conducted,and the results demonstrated the effectiveness of this approach.The sensing and shape prediction steps can run in real-time at a frequency of50 Hz on a consumer-level computer.Additionally,a method is proposed to enhance the robustness of the learning models using data augmentation to handle unexpected sensor failures.All the methods are efficient,not only for in-plane 2D shape estimation but also for out-of-plane 3D shape estimation.The aim of this study is to introduce a methodology for the proprioception of soft pneumatic actuators,including manufacturing and sensing modeling,that can be generalized to other soft robots.展开更多
Humans excel at dexterous manipulation;however,achieving human-level dexterity remains a significant challenge for robots.Technological breakthroughs in the design of anthropomorphic robotic hands,as well as advanceme...Humans excel at dexterous manipulation;however,achieving human-level dexterity remains a significant challenge for robots.Technological breakthroughs in the design of anthropomorphic robotic hands,as well as advancements in visual and tactile perception,have demonstrated significant advantages in addressing this issue.However,coping with the inevitable uncertainty caused by unstructured and dynamic environments in human-like dexterous manipulation tasks,especially for anthropomorphic five-fingered hands,remains an open problem.In this paper,we present a focused review of human-like dexterous manipulation for anthropomorphic five-fingered hands.We begin by defining human-like dexterity and outlining the tasks associated with human-like robot dexterous manipulation.Subsequently,we delve into anthropomorphism and anthropomorphic five-fingered hands,covering definitions,robotic design,and evaluation criteria.Furthermore,we review the learning methods for achieving human-like dexterity in anthropomorphic five-fingered hands,including imitation learning,reinforcement learning and their integration.Finally,we discuss the existing challenges and propose future research directions.This review aims to stimulate interest in scientific research and future applications.展开更多
Robot-assisted microsurgery(RAMS)has many benefits compared to traditional microsurgery.Microsurgical platforms with advanced control strategies,high-quality micro-imaging modalities and micro-sensing systems are wort...Robot-assisted microsurgery(RAMS)has many benefits compared to traditional microsurgery.Microsurgical platforms with advanced control strategies,high-quality micro-imaging modalities and micro-sensing systems are worth developing to further enhance the clinical outcomes of RAMS.Within only a few decades,microsurgical robotics has evolved into a rapidly developing research field with increasing attention all over the world.Despite the appreciated benefits,significant challenges remain to be solved.In this review paper,the emerging concepts and achievements of RAMS will be presented.We introduce the development tendency of RAMS from teleoperation to autonomous systems.We highlight the upcoming new research opportunities that require joint efforts from both clinicians and engineers to pursue further outcomes for RAMS in years to come.展开更多
Bounding is one of the important gaits in quadrupedal locomotion for negotiating obstacles.The authors proposed an effective approach that can learn robust bounding gaits more efficiently despite its large variation i...Bounding is one of the important gaits in quadrupedal locomotion for negotiating obstacles.The authors proposed an effective approach that can learn robust bounding gaits more efficiently despite its large variation in dynamic body movements.The authors first pretrained the neural network(NN)based on data from a robot operated by conventional model-based controllers,and then further optimised the pretrained NN via deep reinforcement learning(DRL).In particular,the authors designed a reward function considering contact points and phases to enforce the gait symmetry and periodicity,which improved the bounding performance.The NN-based feedback controller was learned in the simulation and directly deployed on the real quadruped robot Jueying Mini successfully.A variety of environments are presented both indoors and outdoors with the authors’approach.The authors’approach shows efficient computing and good locomotion results by the Jueying Mini quadrupedal robot bounding over uneven terrain.The cover image is based on the Research Article Efficient learning of robust quadruped bounding using pretrained neural networks by Zhicheng Wang et al.,https://doi.org/10.1049/csy2.12062.展开更多
文摘Robot learning in unstructured environments has been proved to be an extremely challenging problem, mainly because of many uncertainties always present in the real world. Human beings, on the other hand, seem to cope very well with uncertain and unpredictable environments, often relying on perception-based information. Furthermore, humans beings can also utilize perceptions to guide their learning on those parts of the perception-action space that are actually relevant to the task. Therefore, we conduct a research aimed at improving robot learning through the incorporation of both perception-based and measurement-based information. For this reason, a fuzzy reinforcement learning (FRL) agent is proposed in this paper. Based on a neural-fuzzy architecture, different kinds of information can be incorporated into the FRL agent to initialise its action network, critic network and evaluation feedback module so as to accelerate its learning. By making use of the global optimisation capability of GAs (genetic algorithms), a GA-based FRL (GAFRL) agent is presented to solve the local minima problem in traditional actor-critic reinforcement learning. On the other hand, with the prediction capability of the critic network, GAs can perform a more effective global search. Different GAFRL agents are constructed and verified by using the simulation model of a physical biped robot. The simulation analysis shows that the biped learning rate for dynamic balance can be improved by incorporating perception-based information on biped balancing and walking evaluation. The biped robot can find its application in ocean exploration, detection or sea rescue activity, as well as military maritime activity.
基金jointly funded by the National Science and Technology Major Project of the Ministry of Science and Technology of China(2018AAA0102900)the"New Generation Artificial Intelligence"Key Field Research and Development Plan of Guangdong Province(2021B0101410002)。
文摘Humans achieve cognitive development through continuous interaction with their environment,enhancing both perception and behavior.However,current robots lack the capacity for human-like action and evolution,posing a bottleneck to improving robotic intelligence.Existing research predominantly models robots as one-way,static mappings from observations to actions,neglecting the dynamic processes of perception and behavior.This paper introduces a novel approach to robot cognitive learning by considering physical properties.We propose a theoretical framework wherein a robot is conceptualized as a three-body physical system comprising a perception-body(P-body),a cognition-body(C-body),and a behavior-body(B-body).Each body engages in physical dynamics and operates within a closed-loop interaction.Significantly,three crucial interactions connect these bodies.The C-body relies on the Pbody's extracted states and reciprocally offers long-term rewards,optimizing the P-body's perception policy.In addition,the C-body directs the B-body's actions through sub-goals,and subsequent P-body-derived states facilitate the C-body's cognition dynamics learning.At last,the B-body would follow the sub-goal generated by the C-body and perform actions conditioned on the perceptive state from the P-body,which leads to the next interactive step.These interactions foster the joint evolution of each body,culminating in optimal design.To validate our approach,we employ a navigation task using a four-legged robot,D'Kitty,equipped with a movable global camera.Navigational prowess demands intricate coordination of sensing,planning,and D'Kitty's motion.Leveraging our framework yields superior task performance compared with conventional methodologies.In conclusion,this paper establishes a paradigm shift in robot cognitive learning by integrating physical interactions across the P-body,C-body,and B-body,while considering physical properties.Our framework's successful application to a navigation task underscores its efficacy in enhancing robotic intelligence.
基金supported by National Natural Science Foundation of China (No.6120333661273338 and 61003014)Major State Basic Research Development Program of China (973 Program)(No.2013CB329502)
文摘The skill of robotic hand-eye coordination not only helps robots to deal with real time environment,but also afects the fundamental framework of robotic cognition.A number of approaches have been developed in the literature for construction of the robotic hand-eye coordination.However,several important features within infant developmental procedure have not been introduced into such approaches.This paper proposes a new method for robotic hand-eye coordination by imitating the developmental progress of human infants.The work employs a brain-like neural network system inspired by infant brain structure to learn hand-eye coordination,and adopts a developmental mechanism from psychology to drive the robot.The entire learning procedure is driven by developmental constraint： The robot starts to act under fully constrained conditions,when the robot learning system becomes stable,a new constraint is assigned to the robot.After that,the robot needs to act with this new condition again.When all the contained conditions have been overcome,the robot is able to obtain hand-eye coordination ability.The work is supported by experimental evaluation,which shows that the new approach is able to drive the robot to learn autonomously,and make the robot also exhibit developmental progress similar to human infants.
文摘Collaborative heterogeneous robot systems can greatly enhance the efficiency of target search and navigation tasks.In this paper,we design a heterogeneous robot system consisting of an unmanned aerial vehicle(UAV)and an unmanned ground vehicle(UGV)for search and rescue missions in unknown environments.The system is able to search for targets and navigate to them in a maze-like mine environment with the policies learned through deep reinforcement learning algorithms.During the training process,if two robots are trained simultaneously,the rewards related to their collaboration may not be properly obtained.Hence,we introduce a multi-stage reinforcement learning framework and a curiosity module to encourage agents to explore unvisited environments.Experiments in simulation environments show that our framework can train the heterogeneous robot system to achieve the search and navigation with unknown target locations while existing baselines may not.The UGV achieves a success rate of 89.1%in the mission within the original environment,and maintains a 67.6%success rate in untrained complex environments.
基金Supported by National Natural Science Foundation of China(Grant No.52475280)Shaanxi Provincial Natural Science Basic Research Program(Grant No.2025SYSSYSZD-105).
文摘Robots are key to expanding the scope of space applications.The end-to-end training for robot vision-based detection and precision operations is challenging owing to constraints such as extreme environments and high computational overhead.This study proposes a lightweight integrated framework for grasp detection and imitation learning,named GD-IL;it comprises a grasp detection algorithm based on manipulability and Gaussian mixture model(manipulability-GMM),and a grasp trajectory generation algorithm based on a two-stage robot imitation learning algorithm(TS-RIL).In the manipulability-GMM algorithm,we apply GMM clustering and ellipse regression to the object point cloud,propose two judgment criteria to generate multiple candidate grasp bounding boxes for the robot,and use manipulability as a metric for selecting the optimal grasp bounding box.The stages of the TS-RIL algorithm are grasp trajectory learning and robot pose optimization.In the first stage,the robot grasp trajectory is characterized using a second-order dynamic movement primitive model and Gaussian mixture regression(GMM).By adjusting the function form of the forcing term,the robot closely approximates the target-grasping trajectory.In the second stage,a robot pose optimization model is built based on the derived pose error formula and manipulability metric.This model allows the robot to adjust its configuration in real time while grasping,thereby effectively avoiding singularities.Finally,an algorithm verification platform is developed based on a Robot Operating System and a series of comparative experiments are conducted in real-world scenarios.The experimental results demonstrate that GD-IL significantly improves the effectiveness and robustness of grasp detection and trajectory imitation learning,outperforming existing state-of-the-art methods in execution efficiency,manipulability,and success rate.
基金National Natural Science Foundation of China(No.61773374)National Key Research and Development Program of China(No.2017YFB1300104).
文摘Endowing quadruped robots with the skill to forward jump is conducive to making it overcome barriers and pass through complex terrains.In this paper,a model-free control architecture with target-guided policy optimization and deep reinforcement learn-ing(DRL)for quadruped robot jumping is presented.First,the jumping phase is divided into take-off and flight-landing phases,and op-timal strategies with soft actor-critic(SAC)are constructed for the two phases respectively.Second,policy learning including expecta-tions,penalties in the overall jumping process,and extrinsic excitations is designed.Corresponding policies and constraints are all provided for successful take-off,excellent flight attitude and stable standing after landing.In order to avoid low efficiency of random ex-ploration,a curiosity module is introduced as extrinsic rewards to solve this problem.Additionally,the target-guided module encour-ages the robot explore closer and closer to desired jumping target.Simulation results indicate that the quadruped robot can realize com-pleted forward jumping locomotion with good horizontal and vertical distances,as well as excellent motion attitudes.
基金This work was partially supported by Consejo Nacional de Humanidades,Ciencias y Tecnologías(CONAHCyT)the Engineering and Physical Sciences Research Council(grant No.EP/X018962/1).
文摘Bagging is an essential skill that humans perform in their daily activities.However,deformable objects,such as bags,are complex for robots to manipulate.A learning-based framework that enables robots to learn bagging is presented.The novelty of this framework is its ability to learn and perform bagging without relying on simulations.The learning process is accomplished through a reinforcement learning(RL)algorithm introduced and designed to find the best grasping points of the bag based on a set of compact state representations.The framework utilises a set of primitive actions and represents the task in five states.In our experiments,the framework reached 60% and 80% success rates after around 3 h of training in the real world when starting the bagging task from folded and unfolded states,respectively.Finally,the authors test the trained RL model with eight more bags of different sizes to evaluate its generalisability.
基金supported by the European Research Council′s(ERC)starting grant Ergo-Lean(No.GA 850932)funding provided by The Chinese University of Hong Kong,China.
文摘Reactive planning and control capacity for collaborative robots is essential when the tasks change online in an unstructured environment.This is more difficult for collaborative mobile manipulators(CMM)due to high redundancies.To this end,this paper proposed a reactive whole-body locomotion-integrated manipulation approach based on combined learning and optimization.First,human demonstrations are collected,where the wrist and pelvis movements are treated as whole-body trajectories,mapping to the end-effector(EE)and the mobile base(MB)of CMM,respectively.A time-input kernelized movement primitive(T-KMP)learns the whole-body trajectory,and a multi-dimensional kernelized movement primitive(M-KMP)learns the spatial relationship between the MB and EE pose.According to task changes,the T-KMP adapts the learned trajectories online by inserting the new desired point predicted by MKMP.Then,the updated reference trajectories are sent to a hierarchical quadratic programming(HQP)controller,where the EE and the MB trajectories tracking are set as the first and second priority tasks,generating the feasible and optimal joint level commands.An ablation simulation experiment with CMM of the HQP is conducted to show the necessity of MB trajectory tracking in mimicking human whole-body motion behavior.Finally,the tasks of the reactive pick-and-place and reactive reaching were undertaken,where the target object was randomly moved,even out of the region of demonstrations.The results showed that the proposed approach can successfully transfer and adapt the human whole-body loco-manipulation skills to CMM online with task changes.
基金National Natural Science Foundation of China(Nos.62225304,92148204 and 62061160371)National Key Research and Development Program of China(Nos.2021ZD0114503 and 2019YFB1703600)Beijing Top Discipline for Artificial Intelligence Science and Engineering,University of Science and Technology Beijing,and the Beijing Natural Science Foundation(No.JQ20026).
文摘In this article,a robot skills learning framework is developed,which considers both motion modeling and execution.In order to enable the robot to learn skills from demonstrations,a learning method called dynamic movement primitives(DMPs)is introduced to model motion.A staged teaching strategy is integrated into DMPs frameworks to enhance the generality such that the complicated tasks can be also performed for multi-joint manipulators.The DMP connection method is used to make an accurate and smooth transition in position and velocity space to connect complex motion sequences.In addition,motions are categorized into different goals and durations.It is worth mentioning that an adaptive neural networks(NNs)control method is proposed to achieve highly accurate trajectory tracking and to ensure the performance of action execution,which is beneficial to the improvement of reliability of the skills learning system.The experiment test on the Baxter robot verifies the effectiveness of the proposed method.
基金supported by International Cooperation Program of the Natural Science Foundation of China(Grant No.52261135542)Zhejiang Provincial Natural Science Foundation of China(Grant No.LD22E050002)+1 种基金Zhejiang University Global Partnership Fundgrateful to the Russian Science Foundation(Grant No.23-43-00057)for financial support。
文摘Real-time proprioception presents a significant challenge for soft robots due to their infinite degrees of freedom and intrinsic compliance.Previous studies mostly focused on specific sensors and actuators.There is still a lack of generalizable technologies for integrating soft sensing elements into soft actuators and mapping sensor signals to proprioception parameters.To tackle this problem,we employed multi-material 3D printing technology to fabricate sensorized soft-bending actuators(SBAs)using plain and conductive thermoplastic polyurethane(TPU)filaments.We designed various geometric shapes for the sensors and investigated their strain-resistive performance during deformation.To address the nonlinear time-variant behavior of the sensors during dynamic modeling,we adopted a data-driven approach using different deep neural networks to learn the relationship between sensor signals and system states.A series of experiments in various actuation scenarios were conducted,and the results demonstrated the effectiveness of this approach.The sensing and shape prediction steps can run in real-time at a frequency of50 Hz on a consumer-level computer.Additionally,a method is proposed to enhance the robustness of the learning models using data augmentation to handle unexpected sensor failures.All the methods are efficient,not only for in-plane 2D shape estimation but also for out-of-plane 3D shape estimation.The aim of this study is to introduce a methodology for the proprioception of soft pneumatic actuators,including manufacturing and sensing modeling,that can be generalized to other soft robots.
基金supported in part by the National Natural Science Foundation of China(91748131,62006229,and 61771471)in part by Young Scientists Fund of the National Natural Science Foundation of China(62303454)+1 种基金in part by the Strategic Priority Research Program of Chinese Academy of Science(XDB32050106)in part by the InnoHK Project.
文摘Humans excel at dexterous manipulation;however,achieving human-level dexterity remains a significant challenge for robots.Technological breakthroughs in the design of anthropomorphic robotic hands,as well as advancements in visual and tactile perception,have demonstrated significant advantages in addressing this issue.However,coping with the inevitable uncertainty caused by unstructured and dynamic environments in human-like dexterous manipulation tasks,especially for anthropomorphic five-fingered hands,remains an open problem.In this paper,we present a focused review of human-like dexterous manipulation for anthropomorphic five-fingered hands.We begin by defining human-like dexterity and outlining the tasks associated with human-like robot dexterous manipulation.Subsequently,we delve into anthropomorphism and anthropomorphic five-fingered hands,covering definitions,robotic design,and evaluation criteria.Furthermore,we review the learning methods for achieving human-like dexterity in anthropomorphic five-fingered hands,including imitation learning,reinforcement learning and their integration.Finally,we discuss the existing challenges and propose future research directions.This review aims to stimulate interest in scientific research and future applications.
基金supported by Royal Society Research,UK (No.RGSR1221122)
文摘Robot-assisted microsurgery(RAMS)has many benefits compared to traditional microsurgery.Microsurgical platforms with advanced control strategies,high-quality micro-imaging modalities and micro-sensing systems are worth developing to further enhance the clinical outcomes of RAMS.Within only a few decades,microsurgical robotics has evolved into a rapidly developing research field with increasing attention all over the world.Despite the appreciated benefits,significant challenges remain to be solved.In this review paper,the emerging concepts and achievements of RAMS will be presented.We introduce the development tendency of RAMS from teleoperation to autonomous systems.We highlight the upcoming new research opportunities that require joint efforts from both clinicians and engineers to pursue further outcomes for RAMS in years to come.
基金Key Research Project of Zhejiang Lab,Grant/Award Number:2021NB0AL03Key R&D Program of China,Grant/Award Number:2020YFB1313300。
文摘Bounding is one of the important gaits in quadrupedal locomotion for negotiating obstacles.The authors proposed an effective approach that can learn robust bounding gaits more efficiently despite its large variation in dynamic body movements.The authors first pretrained the neural network(NN)based on data from a robot operated by conventional model-based controllers,and then further optimised the pretrained NN via deep reinforcement learning(DRL).In particular,the authors designed a reward function considering contact points and phases to enforce the gait symmetry and periodicity,which improved the bounding performance.The NN-based feedback controller was learned in the simulation and directly deployed on the real quadruped robot Jueying Mini successfully.A variety of environments are presented both indoors and outdoors with the authors’approach.The authors’approach shows efficient computing and good locomotion results by the Jueying Mini quadrupedal robot bounding over uneven terrain.The cover image is based on the Research Article Efficient learning of robust quadruped bounding using pretrained neural networks by Zhicheng Wang et al.,https://doi.org/10.1049/csy2.12062.
