Tires are integral to vehicular systems,directly influencing both safety and overall performance.Tradi-tional tire pressure inspection methods—such as manual or gauge-based approaches—are often time-consuming,prone ...Tires are integral to vehicular systems,directly influencing both safety and overall performance.Tradi-tional tire pressure inspection methods—such as manual or gauge-based approaches—are often time-consuming,prone to inconsistency,and lack the flexibility needed to meet diverse operational demands.In this research,we introduce an AI-driven tire pressure detection system that leverages an enhanced GoogLeNet architecture incorporating a novel Softplus-LReLU activation function.By combining the smooth,non-saturating characteristics of Softplus with a linear adjustment term,this activation function improves computational efficiency and helps stabilize network gradients,thereby mitigating issues such as gradient vanishing and neuron death.Our enhanced GoogLeNet algorithm was validated on a dedicated tire pressure image database comprising three categories-low pressure,normal pressure,and undetected.Experimental results revealed a classification accuracy of 98.518%within 11 min and 56 s of total processing time,substantially surpassing the original GoogLeNet’s 95.1852%and ResNet18’s 92.7778%.This performance gain is attributed to superior feature extraction within the Inception modules and the robust integration of our novel activation function,leading to improved detection reliability and faster inference.Beyond accuracy and speed,the proposed system offers significant benefits for real-time monitoring and vehicle safety by providing timely and precise tire pressure assessments.The automation facilitated by our AI-based method addresses the limitations of manual inspection,delivering consistent,high-quality results that can be easily scaled or customized for various vehicular platforms.Overall,this work establishes a solid foundation for advanced tire pressure monitoring systems and opens avenues for further exploration in AI-assisted vehicle maintenance,contributing to safer and more efficient automotive operations.展开更多
In order to analyze the experimental cardiovascular signal with high accuracy, a system, integrating real-time monitoring and off-line further analysis, was developed and verified. The design, data processing and anal...In order to analyze the experimental cardiovascular signal with high accuracy, a system, integrating real-time monitoring and off-line further analysis, was developed and verified. The design, data processing and analysis methods as well as testing results are described. With 5 sampling frequency choices and 8 channel data acquisition, the system achieved high performances in beat-to-beat monitoring, signal processing and analysis. Tests were carried out to validate its performance in real-time monitoring, effectiveness of digital filters, QRS and blood pressure detection reliability, and RR-interval timing accuracy. The QRS detection rate was at least 99.46% for the records with few noises from MIT-BIH arrhythmia database using the algorithm for real-time monitoring, and no less than 96.43% for the records with some noises. In the condition that noise amplitude levels were less than 80%,the standard deviations for RR-interval timing were less than 1 ms with a generated ECG corrupted with various noises from MIT-BIH Noise Stress Test Database. Besides, the system is open for function expansion to meet further study-specific needs.展开更多
Mimicking human skin mechanoreceptors grouped by various thresholds creates an efficient system to detect interfacial stress between skin and environment,enabling precise human perception.Specifically,the detected sig...Mimicking human skin mechanoreceptors grouped by various thresholds creates an efficient system to detect interfacial stress between skin and environment,enabling precise human perception.Specifically,the detected signals are transmitted in the form of spikes in the neuronal network via synapses.However,current efforts replicating this mechanism for healthmonitoring struggle with limitations in flexibility,durability,and performance,particularly in terms of low sensitivity and narrow detection range.This study develops novel soft mechanoreceptors with tunable pressure thresholds from 1.94 kPa to 15 MPa.The 0.455-mm-thin mechanoreceptor achieves an impressive on–off ratio of over eight orders of magnitude,up to 40,000 repeated compression cycles and after 20 wash cycles.In addition,the helical array reduces the complexity and port count,requiring only two output channels,and a differential simplification algorithm enables two-dimensional spatial mapping of pressure.This array shows stable performance across temperatures ranging from−40 to 50°C and underwater at depths of 1 m.This technology shows significant potential for wearable healthcare applications,including sensor stimulation for children and the elderly,and fall detection for Parkinson’s patients,thereby enhancing the functionality and reliability of wearable monitoring systems.展开更多
Although the deuterium and helium have almost the same mass,a Penning Optical Gas Analyzer（POGA） system on the basis of the spectroscopic method and Penning discharging has been designed on EAST,since 2014.The POGA ...Although the deuterium and helium have almost the same mass,a Penning Optical Gas Analyzer（POGA） system on the basis of the spectroscopic method and Penning discharging has been designed on EAST,since 2014.The POGA system was developed successfully in 2015,it was the first time that EAST could detect helium partial pressure in deuterium plasma（wall conditioning and plasma operation scenario）.With dedicated calibration and proper adjustment of the parameters,the minimum concentration of helium in deuterium gas can be measured as about 0.5% instead of 1% on the other tokamak devices.Moreover,the He and D2 partial pressures are measured simultaneously.At present,the measurable range of deuterium partial pressure is 1×10^-7 mbar to 1×10^-5mbar,meanwhile the range of helium is 1×10^-8 mbar to 1×10^-5 mbar.The measurable range can be modified by means of the adjustment of POGA system＇s parameters.It is possible to detect the interesting part of the gas with a time resolution of less than 5 ms（the 200 ms because of conductance of transfer pipe at present）.The POGA system was routinely employed to wall conditioning and helium enrichment investigation in2015.Last but not the least,the low temperature plasma of POGA is generated by normal penning gauge Pfeiffer IKR gauge instead of Alcatel CF2 P,which has been suspended for a few years and was used for almost all the POGA systems in the world.展开更多
Scanning the absorption spectral line of water vapor through wavelength around 1368.597nm is successfully used to measure the value of micro-moisture content. The synchronous superposition average of original signal a...Scanning the absorption spectral line of water vapor through wavelength around 1368.597nm is successfully used to measure the value of micro-moisture content. The synchronous superposition average of original signal algorithm based on labview is innovated and applied to detecting weak spectrum absorption signal instead of low pass filter. Two data processing methods are used to get the concentration of water vapor in ppm： one is a general formula method which has newly deduced a general formula to calculate the concentration of gas with temperature and beam intensity ratio when the pressure is equal to or greater than 1 atm; the other is engineering calibration method which is proved to have high resolution and accuracy with the fitted curve of beam intensity ratio and concentration in ppm when the temperature changes form 258K to 305K and the pressure ranges from 1 atm to 5 atm.展开更多
In a recent study published in Cell,Xie et al.1 unveiled the mechanisms by which the 500 million enteric neurons in the enteric nervous system(ENS)sense and respond to mechanical forces.Their research indicates that P...In a recent study published in Cell,Xie et al.1 unveiled the mechanisms by which the 500 million enteric neurons in the enteric nervous system(ENS)sense and respond to mechanical forces.Their research indicates that Piezo1+cholinergic excitatory neurons are key mechanosensors for detecting intestinal pressure and are crucial for maintaining normal gut motility and inflammatory homeostasis.展开更多
Compared with piezoresistive sensors,pressure sensors based on the contact resistance effect are proven to have higher sensitivity and the ability to detect ultra-low pressure,thus attracting extensive research intere...Compared with piezoresistive sensors,pressure sensors based on the contact resistance effect are proven to have higher sensitivity and the ability to detect ultra-low pressure,thus attracting extensive research interest in wearable devices and artificial intelligence systems.However,most studies focus on static or low-frequency pressure detection,and there are few reports on high-frequency dynamic pressure detection.Limited by the viscoelasticity of polymers(necessary materials for traditional vibration sensors),the development of vibration sensors with high frequency response remains a great challenge.Here,we report a graphene aerogel-based vibration sensor with higher sensitivity and wider frequency response range(2 Hz–10 kHz)than both conventional piezoresistive and similar sensors.By modulating the microscopic morphology and mechanical properties,the super-elastic graphene aerogels suitable for vibration sensing have been prepared successfully.Meanwhile,the mechanism of the effect of density on the vibration sensor’s sensitivity is studied in detail.On this basis,the sensitivity,signal fidelity and signal-to-noise ratio of the sensor are further improved by optimizing the structure configuration.The developed sensor exhibits remarkable repeatability,excellent stability,high resolution(0.0039 g)and good linearity(non-linearity error<0.8%)without hysteresis.As demos,the sensor can not only monitor low-frequency physiological signals and motion of the human body,but also respond to the high-frequency vibrations of rotating machines.In addition,the sensor can also detect static pressure.We expect the vibration sensor to meet a wider range of functional needs in wearable devices,smart robots,and industrial equipment.展开更多
基金the National Science and Technology Council,Taiwan,for financially supporting this research(Grant No.NSTC 113-2221-E-018-011)Ministry of Education’s Teaching Practice Research Program,Taiwan(PSK1134099).
文摘Tires are integral to vehicular systems,directly influencing both safety and overall performance.Tradi-tional tire pressure inspection methods—such as manual or gauge-based approaches—are often time-consuming,prone to inconsistency,and lack the flexibility needed to meet diverse operational demands.In this research,we introduce an AI-driven tire pressure detection system that leverages an enhanced GoogLeNet architecture incorporating a novel Softplus-LReLU activation function.By combining the smooth,non-saturating characteristics of Softplus with a linear adjustment term,this activation function improves computational efficiency and helps stabilize network gradients,thereby mitigating issues such as gradient vanishing and neuron death.Our enhanced GoogLeNet algorithm was validated on a dedicated tire pressure image database comprising three categories-low pressure,normal pressure,and undetected.Experimental results revealed a classification accuracy of 98.518%within 11 min and 56 s of total processing time,substantially surpassing the original GoogLeNet’s 95.1852%and ResNet18’s 92.7778%.This performance gain is attributed to superior feature extraction within the Inception modules and the robust integration of our novel activation function,leading to improved detection reliability and faster inference.Beyond accuracy and speed,the proposed system offers significant benefits for real-time monitoring and vehicle safety by providing timely and precise tire pressure assessments.The automation facilitated by our AI-based method addresses the limitations of manual inspection,delivering consistent,high-quality results that can be easily scaled or customized for various vehicular platforms.Overall,this work establishes a solid foundation for advanced tire pressure monitoring systems and opens avenues for further exploration in AI-assisted vehicle maintenance,contributing to safer and more efficient automotive operations.
基金This work is supported by Beijing Natural Science Foundation (3052015)
文摘In order to analyze the experimental cardiovascular signal with high accuracy, a system, integrating real-time monitoring and off-line further analysis, was developed and verified. The design, data processing and analysis methods as well as testing results are described. With 5 sampling frequency choices and 8 channel data acquisition, the system achieved high performances in beat-to-beat monitoring, signal processing and analysis. Tests were carried out to validate its performance in real-time monitoring, effectiveness of digital filters, QRS and blood pressure detection reliability, and RR-interval timing accuracy. The QRS detection rate was at least 99.46% for the records with few noises from MIT-BIH arrhythmia database using the algorithm for real-time monitoring, and no less than 96.43% for the records with some noises. In the condition that noise amplitude levels were less than 80%,the standard deviations for RR-interval timing were less than 1 ms with a generated ECG corrupted with various noises from MIT-BIH Noise Stress Test Database. Besides, the system is open for function expansion to meet further study-specific needs.
基金Grants Council of Hong Kong(Grant No.T42-513/24-R)Innovation and Technology Fund(Grant No.MRP/020/21)The Hong Kong Polytechnic University(Grant No.847A).
文摘Mimicking human skin mechanoreceptors grouped by various thresholds creates an efficient system to detect interfacial stress between skin and environment,enabling precise human perception.Specifically,the detected signals are transmitted in the form of spikes in the neuronal network via synapses.However,current efforts replicating this mechanism for healthmonitoring struggle with limitations in flexibility,durability,and performance,particularly in terms of low sensitivity and narrow detection range.This study develops novel soft mechanoreceptors with tunable pressure thresholds from 1.94 kPa to 15 MPa.The 0.455-mm-thin mechanoreceptor achieves an impressive on–off ratio of over eight orders of magnitude,up to 40,000 repeated compression cycles and after 20 wash cycles.In addition,the helical array reduces the complexity and port count,requiring only two output channels,and a differential simplification algorithm enables two-dimensional spatial mapping of pressure.This array shows stable performance across temperatures ranging from−40 to 50°C and underwater at depths of 1 m.This technology shows significant potential for wearable healthcare applications,including sensor stimulation for children and the elderly,and fall detection for Parkinson’s patients,thereby enhancing the functionality and reliability of wearable monitoring systems.
基金funded by National Magnetic Confinement Fusion Science Program of China under Contract No.2013GB114004,No.2014GB106005 & No.2015GB101000National Nature Science Foundation of China under Contract No.11625524,No.11321092 and No.11405210partly supported by the Japan Society for the Promotion of ScienceNational Research Foundation of Korea-National Science Foundation of China(JSPS-NRF-NSFC) A3 Foresight Program in the field of Plasma Physics(NSFC No.11261140328)
文摘Although the deuterium and helium have almost the same mass,a Penning Optical Gas Analyzer（POGA） system on the basis of the spectroscopic method and Penning discharging has been designed on EAST,since 2014.The POGA system was developed successfully in 2015,it was the first time that EAST could detect helium partial pressure in deuterium plasma（wall conditioning and plasma operation scenario）.With dedicated calibration and proper adjustment of the parameters,the minimum concentration of helium in deuterium gas can be measured as about 0.5% instead of 1% on the other tokamak devices.Moreover,the He and D2 partial pressures are measured simultaneously.At present,the measurable range of deuterium partial pressure is 1×10^-7 mbar to 1×10^-5mbar,meanwhile the range of helium is 1×10^-8 mbar to 1×10^-5 mbar.The measurable range can be modified by means of the adjustment of POGA system＇s parameters.It is possible to detect the interesting part of the gas with a time resolution of less than 5 ms（the 200 ms because of conductance of transfer pipe at present）.The POGA system was routinely employed to wall conditioning and helium enrichment investigation in2015.Last but not the least,the low temperature plasma of POGA is generated by normal penning gauge Pfeiffer IKR gauge instead of Alcatel CF2 P,which has been suspended for a few years and was used for almost all the POGA systems in the world.
基金This work was supported by Natural Science Foundation of China （60977058）, Science Fund for Distinguished Young Scholars of Shandong Province of China （JQ200819）, Research Award Fund for Outstanding Middle-aged＇ and Young Scientist of Shandong Province of China （2007BS08003）, Independent Innovation Foundation of Shandong University （IIFSDU2010JC002）.
文摘Scanning the absorption spectral line of water vapor through wavelength around 1368.597nm is successfully used to measure the value of micro-moisture content. The synchronous superposition average of original signal algorithm based on labview is innovated and applied to detecting weak spectrum absorption signal instead of low pass filter. Two data processing methods are used to get the concentration of water vapor in ppm： one is a general formula method which has newly deduced a general formula to calculate the concentration of gas with temperature and beam intensity ratio when the pressure is equal to or greater than 1 atm; the other is engineering calibration method which is proved to have high resolution and accuracy with the fitted curve of beam intensity ratio and concentration in ppm when the temperature changes form 258K to 305K and the pressure ranges from 1 atm to 5 atm.
基金supported by the Zhejiang Provincial Natural Science Foundation of China(Y24H160082)the Medical Interdisciplinary Innovation Program 2024,Zhejiang University School of Medicine.
文摘In a recent study published in Cell,Xie et al.1 unveiled the mechanisms by which the 500 million enteric neurons in the enteric nervous system(ENS)sense and respond to mechanical forces.Their research indicates that Piezo1+cholinergic excitatory neurons are key mechanosensors for detecting intestinal pressure and are crucial for maintaining normal gut motility and inflammatory homeostasis.
基金supported by the National Key R&D Program of China(Nos.2018YFA0208402 and 2020YFA0714700)the National Natural Science Foundation of China(Nos.52172060,51820105002,11634014 and 51372269)+1 种基金Prof.X.J.W.thanks Youth Innovation Promotion Association of the Chinese Academy of Sciences(No.2020005)One Hundred Talent Project of Institute of Physics,CAS.Prof.H.P.L.and Prof.X.Z.thank support by the“One Hundred talents project”of CAS.
文摘Compared with piezoresistive sensors,pressure sensors based on the contact resistance effect are proven to have higher sensitivity and the ability to detect ultra-low pressure,thus attracting extensive research interest in wearable devices and artificial intelligence systems.However,most studies focus on static or low-frequency pressure detection,and there are few reports on high-frequency dynamic pressure detection.Limited by the viscoelasticity of polymers(necessary materials for traditional vibration sensors),the development of vibration sensors with high frequency response remains a great challenge.Here,we report a graphene aerogel-based vibration sensor with higher sensitivity and wider frequency response range(2 Hz–10 kHz)than both conventional piezoresistive and similar sensors.By modulating the microscopic morphology and mechanical properties,the super-elastic graphene aerogels suitable for vibration sensing have been prepared successfully.Meanwhile,the mechanism of the effect of density on the vibration sensor’s sensitivity is studied in detail.On this basis,the sensitivity,signal fidelity and signal-to-noise ratio of the sensor are further improved by optimizing the structure configuration.The developed sensor exhibits remarkable repeatability,excellent stability,high resolution(0.0039 g)and good linearity(non-linearity error<0.8%)without hysteresis.As demos,the sensor can not only monitor low-frequency physiological signals and motion of the human body,but also respond to the high-frequency vibrations of rotating machines.In addition,the sensor can also detect static pressure.We expect the vibration sensor to meet a wider range of functional needs in wearable devices,smart robots,and industrial equipment.
