In recent decades,capacitive pressure sensors(CPSs)with high sensitivity have demonstrated significant potential in applications such as medical monitoring,artificial intelligence,and soft robotics.Efforts to enhance ...In recent decades,capacitive pressure sensors(CPSs)with high sensitivity have demonstrated significant potential in applications such as medical monitoring,artificial intelligence,and soft robotics.Efforts to enhance this sensitivity have predominantly focused on material design and structural optimization,with surface microstructures such as wrinkles,pyramids,and micro-pillars proving effective.Although finite element modeling(FEM)has guided enhancements in CPS sensitivity across various surface designs,a theoretical understanding of sensitivity improvements remains underexplored.This paper employs sinusoidal wavy surfaces as a representative model to analytically elucidate the underlying mechanisms of sensitivity enhancement through contact mechanics.These theoretical insights are corroborated by FEM and experimental validations.Our findings underscore that optimizing material properties,such as Young’s modulus and relative permittivity,alongside adjustments in surface roughness and substrate thickness,can significantly elevate the sensitivity.The optimal performance is achieved when the amplitude-to-wavelength ratio(H/)is about 0.2.These results offer critical insights for designing ultrasensitive CPS devices,paving the way for advancements in sensor technology.展开更多
针对样机设计领域存在的多源异构知识分散、重用效率低以及研发过程智能化水平不足的问题,本研究提出基于知识图谱与智能推理的辅助设计系统。通过构建面向样机设计领域的专业知识图谱,实现了对设计方案、工艺文档等结构化与非结构化数...针对样机设计领域存在的多源异构知识分散、重用效率低以及研发过程智能化水平不足的问题,本研究提出基于知识图谱与智能推理的辅助设计系统。通过构建面向样机设计领域的专业知识图谱,实现了对设计方案、工艺文档等结构化与非结构化数据的语义化整合。本文的创新性在于提出了面向样机设计领域的专用辅助系统,采用知识图谱驱动模式实现设计要素的智能关联与可视化;同时开发了知识更新模块,实现知识的动态更新迭代;提出基于用户交互行为的动态推荐机制,并且融合自然语言理解(Natural Language Understanding, NLU)与大语言模型技术(Large Language Model,LLM),构建了面向样机研制全生命周期的智能问答模块,有效提升了复杂技术问题的解决效率。文章系统阐述了领域知识图谱构建方法、系统架构设计以及推理推荐算法实现路径,并通过典型装备样机的研制过程验证了系统在知识复用率提升和设计周期缩短等方面的应用价值。研究成果为装备制造企业实现知识驱动的智能化设计提供了可复用的技术框架与实践参考。展开更多
All-inorganic perovskites based on cesium-lead-bromine(Cs-Pb-Br)have been a prominent research focus in optoelectronics in recent years.The optimisation and tunability of their macroscopic properties exploit the confo...All-inorganic perovskites based on cesium-lead-bromine(Cs-Pb-Br)have been a prominent research focus in optoelectronics in recent years.The optimisation and tunability of their macroscopic properties exploit the conformational flexibility,resulting in various crystal structures.Varying synthesis parameters can yield distinct crystal structures from Cs,Pb,and Br precursors,and manually exploring the relationship between these synthesis parameters and the resulting crystal structure is both labour-intensive and time-consuming.Machine learning(ML)can rapidly uncover insights and drive discoveries in chemical synthesis with the support of data,significantly reducing both the cost and development cycle of materials.Here,we gathered synthesis parameters from published literature(220 synthesis runs)and implemented eight distinct ML models,including eXtreme Gradient Boosting(XGB),Decision Tree(DT),Support Vector Machine(SVM),Random Forest(RF),Naïve Bayes(NB),Logistic Regression(LR),Gradient Boosting(GB),and K-Nearest(KN)to classify and predict Cs-Pb-Br crystal structures from given synthesis parameters.Validation accuracy,precision,F1 score,recall,and average area under the curve(AUC)are employed to evaluate these ML models.The XGB model exhibited the best performance,achieving a validation accuracy of 0.841.The trained XGB model was subsequently utilised to predict the structure from 10 experimental runs using a randomised set of parameters,achieving a testing accuracy of 0.8.The results indicate that the Cs/Pb molar ratio,reaction time,and the concentration of organic compounds(ligands)play crucial roles in synthesising various crystal structures of Cs-Pb-Br.This study demonstrates a significant decrease in effort required for experimental procedures and builds a foundational basis for predicting crystal structures from synthesis parameters.展开更多
为解决基于关节力矩的双足机器人参数辨识方法辨识精度不高,基于完整的足底力信息和运动捕捉数据的辨识方法对实验条件要求较高的问题,提出基于ZMP(zero moment point)数据的双足机器人惯性参数辨识方法。将理论ZMP与实际ZMP的位置偏差...为解决基于关节力矩的双足机器人参数辨识方法辨识精度不高,基于完整的足底力信息和运动捕捉数据的辨识方法对实验条件要求较高的问题,提出基于ZMP(zero moment point)数据的双足机器人惯性参数辨识方法。将理论ZMP与实际ZMP的位置偏差作为目标函数,考虑参数范围和机器人总质量两类约束条件,建立只使用双足机器人自身传感器采样数据的惯性参数辨识优化模型。针对所建模型无法拆分成线性形式的问题,推导目标函数关于参数矢量的梯度矢量和海塞矩阵,并给出了基于最速下降法和牛顿法的优化求解算法。使用GoRoBoT-II机器人的双足部分,进行腿部杆件的惯性参数辨识实验,将所提出方法得到的辨识结果与传统基于关节力矩的辨识结果进行对比,发现基于ZMP的辨识方法的结果更接近于三维几何建模得到的参数标称值,且理论ZMP与实际ZMP的偏差均值为4.6 mm,小于传统基于力矩辨识方法的12.4 mm,说明所提出的基于ZMP的惯性参数辨识方法能够得到比传统方法更好的结果。展开更多
基金supported by the National Natural Science Foundation of China(Grant No.12272369)the Strategic Priority Research Program of the Chinese Academy of Sciences(Grant No.XDB0620101).
文摘In recent decades,capacitive pressure sensors(CPSs)with high sensitivity have demonstrated significant potential in applications such as medical monitoring,artificial intelligence,and soft robotics.Efforts to enhance this sensitivity have predominantly focused on material design and structural optimization,with surface microstructures such as wrinkles,pyramids,and micro-pillars proving effective.Although finite element modeling(FEM)has guided enhancements in CPS sensitivity across various surface designs,a theoretical understanding of sensitivity improvements remains underexplored.This paper employs sinusoidal wavy surfaces as a representative model to analytically elucidate the underlying mechanisms of sensitivity enhancement through contact mechanics.These theoretical insights are corroborated by FEM and experimental validations.Our findings underscore that optimizing material properties,such as Young’s modulus and relative permittivity,alongside adjustments in surface roughness and substrate thickness,can significantly elevate the sensitivity.The optimal performance is achieved when the amplitude-to-wavelength ratio(H/)is about 0.2.These results offer critical insights for designing ultrasensitive CPS devices,paving the way for advancements in sensor technology.
文摘针对样机设计领域存在的多源异构知识分散、重用效率低以及研发过程智能化水平不足的问题,本研究提出基于知识图谱与智能推理的辅助设计系统。通过构建面向样机设计领域的专业知识图谱,实现了对设计方案、工艺文档等结构化与非结构化数据的语义化整合。本文的创新性在于提出了面向样机设计领域的专用辅助系统,采用知识图谱驱动模式实现设计要素的智能关联与可视化;同时开发了知识更新模块,实现知识的动态更新迭代;提出基于用户交互行为的动态推荐机制,并且融合自然语言理解(Natural Language Understanding, NLU)与大语言模型技术(Large Language Model,LLM),构建了面向样机研制全生命周期的智能问答模块,有效提升了复杂技术问题的解决效率。文章系统阐述了领域知识图谱构建方法、系统架构设计以及推理推荐算法实现路径,并通过典型装备样机的研制过程验证了系统在知识复用率提升和设计周期缩短等方面的应用价值。研究成果为装备制造企业实现知识驱动的智能化设计提供了可复用的技术框架与实践参考。
文摘针对水下自主航行器(autonomous underwater vehicle,AUV)在复杂三维洋流环境中目标跟踪的高维、动态干扰和稀疏回报挑战,提出了一种基于分布式强化学习的水下自主航行器水下三维洋流目标跟踪控制算法。首先,引入真实三维洋流数据,设计动态目标跟踪场景,以准确描述AUV的运动过程;其次,结合对抗深度强化学习网络(dueling deep Q-network,Dueling DQN)结构与分位数回归方法,针对三维洋流环境可能导致Q值过高估计的问题,构建分布强化学习框架,以量化Q值的不确定性,提升策略对动态干扰的适应能力;最后,引入优先经验回放机制,设计约束条件下的奖励函数,优化数据采样策略,加速模型收敛。实验结果表明,相较于深度Q网络(deep Q-network,DQN)、双深度Q网络(double deep Q-network,DDQN)和Dueling DQN,所提算法在复杂洋流环境中表现更优,在收敛速度、目标跟踪精度和鲁棒性方面均取得了显著的进展。
基金the Italian Space Agency(Agenzia Spaziale Italiana,ASI)in the framework of the Research Day“Giornate della Ricerca Spaziale”initiative through the contract ASI N.2023-4-U.0.
文摘All-inorganic perovskites based on cesium-lead-bromine(Cs-Pb-Br)have been a prominent research focus in optoelectronics in recent years.The optimisation and tunability of their macroscopic properties exploit the conformational flexibility,resulting in various crystal structures.Varying synthesis parameters can yield distinct crystal structures from Cs,Pb,and Br precursors,and manually exploring the relationship between these synthesis parameters and the resulting crystal structure is both labour-intensive and time-consuming.Machine learning(ML)can rapidly uncover insights and drive discoveries in chemical synthesis with the support of data,significantly reducing both the cost and development cycle of materials.Here,we gathered synthesis parameters from published literature(220 synthesis runs)and implemented eight distinct ML models,including eXtreme Gradient Boosting(XGB),Decision Tree(DT),Support Vector Machine(SVM),Random Forest(RF),Naïve Bayes(NB),Logistic Regression(LR),Gradient Boosting(GB),and K-Nearest(KN)to classify and predict Cs-Pb-Br crystal structures from given synthesis parameters.Validation accuracy,precision,F1 score,recall,and average area under the curve(AUC)are employed to evaluate these ML models.The XGB model exhibited the best performance,achieving a validation accuracy of 0.841.The trained XGB model was subsequently utilised to predict the structure from 10 experimental runs using a randomised set of parameters,achieving a testing accuracy of 0.8.The results indicate that the Cs/Pb molar ratio,reaction time,and the concentration of organic compounds(ligands)play crucial roles in synthesising various crystal structures of Cs-Pb-Br.This study demonstrates a significant decrease in effort required for experimental procedures and builds a foundational basis for predicting crystal structures from synthesis parameters.
文摘为解决基于关节力矩的双足机器人参数辨识方法辨识精度不高,基于完整的足底力信息和运动捕捉数据的辨识方法对实验条件要求较高的问题,提出基于ZMP(zero moment point)数据的双足机器人惯性参数辨识方法。将理论ZMP与实际ZMP的位置偏差作为目标函数,考虑参数范围和机器人总质量两类约束条件,建立只使用双足机器人自身传感器采样数据的惯性参数辨识优化模型。针对所建模型无法拆分成线性形式的问题,推导目标函数关于参数矢量的梯度矢量和海塞矩阵,并给出了基于最速下降法和牛顿法的优化求解算法。使用GoRoBoT-II机器人的双足部分,进行腿部杆件的惯性参数辨识实验,将所提出方法得到的辨识结果与传统基于关节力矩的辨识结果进行对比,发现基于ZMP的辨识方法的结果更接近于三维几何建模得到的参数标称值,且理论ZMP与实际ZMP的偏差均值为4.6 mm,小于传统基于力矩辨识方法的12.4 mm,说明所提出的基于ZMP的惯性参数辨识方法能够得到比传统方法更好的结果。
