Environmental perception is a key technology for autonomous driving.Owing to the limitations of a single sensor,multiple sensors are often used in practical applications.However,multi-sensor fusion faces some problems...Environmental perception is a key technology for autonomous driving.Owing to the limitations of a single sensor,multiple sensors are often used in practical applications.However,multi-sensor fusion faces some problems,such as the choice of sensors and fusion methods.To solve these issues,we proposed a machine learning-based fusion sensing system that uses a camera and radar,and that can be used in intelligent vehicles.First,the object detection algorithm is used to detect the image obtained by the camera;in sequence,the radar data is preprocessed,coordinate transformation is performed,and a multi-layer perceptron model for correlating the camera detection results with the radar data is proposed.The proposed fusion sensing system was verified by comparative experiments in a real-world environment.The experimental results show that the system can effectively integrate camera and radar data results,and obtain accurate and comprehensive object information in front of intelligent vehicles.展开更多
Speeding is one of the primary contributors to rural road crashes.Self-explaining theory offers a solution to reduce speeding,which suggests that well-designed facility environments(i.e.,road facilities and surroundin...Speeding is one of the primary contributors to rural road crashes.Self-explaining theory offers a solution to reduce speeding,which suggests that well-designed facility environments(i.e.,road facilities and surrounding landscapes)can automatically guide drivers to choose appropriate speeds on different road categories.This study proposes an improved lightweight convolutional neural network(LW-CNN)that includes drivers’visual perception characteristics(i.e.,depth perception and dynamic vision)to conduct the self-explaining analysis of the facility environment on 2-lane rural roads.Data for this study are gathered through naturalistic driving experiments on 2-lane rural roads across five Chinese provinces.A total of 3502 visual facility environment images,alongside their corresponding operation speeds and speed limits,are collected.The improved LW-CNN exhibits high accuracy and efficiency in predicting operation speeds with these visual facility environment images,achieving a train loss of 0.05%and a validation loss of 0.15%.The semantics of facility environments affecting operation speeds are further identified by combining this LW-CNN with the gradient-weighted class activation mapping(Grad-CAM)algorithm and the semantic segmentation network.Then,six typical 2-lane rural road categories perceived by drivers with different operation speeds and speeding probability(SP)are sum-marized using k-means clustering.An objective and comprehensive analysis of each category’s semantic composition and depth features is conducted to evaluate their influence on drivers’speeding probability and road category perception.The findings of this study can be directly used to optimize facility environments from drivers’visual perception to decrease speeding-related crashes.展开更多
针对地下复杂环境中机器人感知系统面临的低光照干扰与计算资源受限双重挑战,提出一种轻量化双模态目标检测方法。通过构建融合激光雷达(LiDAR)点云与RGB图像的双分支网络架构,在浅层、中层和深层实现多尺度特征融合。所提方法引入StarF...针对地下复杂环境中机器人感知系统面临的低光照干扰与计算资源受限双重挑战,提出一种轻量化双模态目标检测方法。通过构建融合激光雷达(LiDAR)点云与RGB图像的双分支网络架构,在浅层、中层和深层实现多尺度特征融合。所提方法引入StarFusion模块,采用逐元素乘法增强跨模态特征交互,结合深度可分离卷积与通道压缩策略,将模型参数量压缩至2.3M。为突破算法验证瓶颈,构建包含4类地下典型目标的低光照多模态数据集,其图像亮度(25±8.3)与清晰度(18.6±6.9)显著低于常规数据集。实验表明,本文方法在自建数据集上mAP50(交并比为0.5时的平均精度均值)达到86.1%,较基准算法YOLOv8提升2.6%,推理速度达20帧/秒。将该方法实际部署于Jetson Orin NX平台的勘探机器人,结果表明,双模态互补机制有效克服了单传感器在低光照环境下的感知盲区,为地下自主作业提供了可靠的实时环境感知解决方案。展开更多
The study of the spallation of thermal barrier coatings on turbine blades and its influence is of great significance for gas turbine safety operation.However,numerical simulation related to thermal barrier coatings is...The study of the spallation of thermal barrier coatings on turbine blades and its influence is of great significance for gas turbine safety operation.However,numerical simulation related to thermal barrier coatings is difficult and time-costly,which makes it hard to meet engineering demands.Therefore,this work establishes a rapid prediction model for the surface temperature and cooling efficiency of turbine blades with localized spallation of thermal barrier coatings based on a thin-wall thermal resistance model.Firstly,the influence of localized spallation of thermal barrier coatings on the cooling efficiency of typical turbine blades is numerically investigated.Then,based on the simulation data set and multi-layer perception(MLP)neural network,an intelligent prediction model for the temperature and cooling efficiency distribution of localized spallation of coatings is constructed,which can rapidly predict the surface temperature and cooling efficiency of the blade under the situation of spallation of coating at any position on the blade surface.The results show that,under a certain spallation area,the shape of localized coating spallation has little influence on the cooling efficiency,while the increase of spallation thickness will cause a linear increase in the average temperature of the blade surface.The prediction error of the proposed rapid prediction model for the average surface temperature and cooling efficiency of blades is within 2%,and the prediction error of the temperature and cooling efficiency at the spallation position is within 6%for 80%of the samples,with an overall average error within 10%.It is concluded from the rapid prediction model that when the depth of coating spallation increases,the closer the spallation position is to the leading edge of the blade,the greater the difference in cooling efficiency is,and the degree of influence of coating spallation on the cooling efficiency also increases.展开更多
基金the National Natural Science Foundation of China(No.U1764264/61873165)the Shanghai Automotive Industry Science and Technology Development Foundation(No.1733/1807)。
文摘Environmental perception is a key technology for autonomous driving.Owing to the limitations of a single sensor,multiple sensors are often used in practical applications.However,multi-sensor fusion faces some problems,such as the choice of sensors and fusion methods.To solve these issues,we proposed a machine learning-based fusion sensing system that uses a camera and radar,and that can be used in intelligent vehicles.First,the object detection algorithm is used to detect the image obtained by the camera;in sequence,the radar data is preprocessed,coordinate transformation is performed,and a multi-layer perceptron model for correlating the camera detection results with the radar data is proposed.The proposed fusion sensing system was verified by comparative experiments in a real-world environment.The experimental results show that the system can effectively integrate camera and radar data results,and obtain accurate and comprehensive object information in front of intelligent vehicles.
基金supported by the National Natural Science Foundation of China(No.52102416)the Natural Science Foundation of Shanghai(No.22ZR1466000)the Fundamental Research Funds for the Central Universities of China(No.22120240159).
文摘Speeding is one of the primary contributors to rural road crashes.Self-explaining theory offers a solution to reduce speeding,which suggests that well-designed facility environments(i.e.,road facilities and surrounding landscapes)can automatically guide drivers to choose appropriate speeds on different road categories.This study proposes an improved lightweight convolutional neural network(LW-CNN)that includes drivers’visual perception characteristics(i.e.,depth perception and dynamic vision)to conduct the self-explaining analysis of the facility environment on 2-lane rural roads.Data for this study are gathered through naturalistic driving experiments on 2-lane rural roads across five Chinese provinces.A total of 3502 visual facility environment images,alongside their corresponding operation speeds and speed limits,are collected.The improved LW-CNN exhibits high accuracy and efficiency in predicting operation speeds with these visual facility environment images,achieving a train loss of 0.05%and a validation loss of 0.15%.The semantics of facility environments affecting operation speeds are further identified by combining this LW-CNN with the gradient-weighted class activation mapping(Grad-CAM)algorithm and the semantic segmentation network.Then,six typical 2-lane rural road categories perceived by drivers with different operation speeds and speeding probability(SP)are sum-marized using k-means clustering.An objective and comprehensive analysis of each category’s semantic composition and depth features is conducted to evaluate their influence on drivers’speeding probability and road category perception.The findings of this study can be directly used to optimize facility environments from drivers’visual perception to decrease speeding-related crashes.
文摘针对地下复杂环境中机器人感知系统面临的低光照干扰与计算资源受限双重挑战,提出一种轻量化双模态目标检测方法。通过构建融合激光雷达(LiDAR)点云与RGB图像的双分支网络架构,在浅层、中层和深层实现多尺度特征融合。所提方法引入StarFusion模块,采用逐元素乘法增强跨模态特征交互,结合深度可分离卷积与通道压缩策略,将模型参数量压缩至2.3M。为突破算法验证瓶颈,构建包含4类地下典型目标的低光照多模态数据集,其图像亮度(25±8.3)与清晰度(18.6±6.9)显著低于常规数据集。实验表明,本文方法在自建数据集上mAP50(交并比为0.5时的平均精度均值)达到86.1%,较基准算法YOLOv8提升2.6%,推理速度达20帧/秒。将该方法实际部署于Jetson Orin NX平台的勘探机器人,结果表明,双模态互补机制有效克服了单传感器在低光照环境下的感知盲区,为地下自主作业提供了可靠的实时环境感知解决方案。
基金supported by the National Natural Science Foundation of China(No.52206090)the Jiangsu Provincial Natural Science Foundation(No.BK20220901)+2 种基金the National Major Science and Technology Projects of China(No.Y2022-Ⅲ-0004-0013)Engineering Research Center of Low-Carbon Aerospace Power Ministry of Education(No.CEPE2024020)the China Postdoctoral Science Foundation(No.2022TQ0149).
文摘The study of the spallation of thermal barrier coatings on turbine blades and its influence is of great significance for gas turbine safety operation.However,numerical simulation related to thermal barrier coatings is difficult and time-costly,which makes it hard to meet engineering demands.Therefore,this work establishes a rapid prediction model for the surface temperature and cooling efficiency of turbine blades with localized spallation of thermal barrier coatings based on a thin-wall thermal resistance model.Firstly,the influence of localized spallation of thermal barrier coatings on the cooling efficiency of typical turbine blades is numerically investigated.Then,based on the simulation data set and multi-layer perception(MLP)neural network,an intelligent prediction model for the temperature and cooling efficiency distribution of localized spallation of coatings is constructed,which can rapidly predict the surface temperature and cooling efficiency of the blade under the situation of spallation of coating at any position on the blade surface.The results show that,under a certain spallation area,the shape of localized coating spallation has little influence on the cooling efficiency,while the increase of spallation thickness will cause a linear increase in the average temperature of the blade surface.The prediction error of the proposed rapid prediction model for the average surface temperature and cooling efficiency of blades is within 2%,and the prediction error of the temperature and cooling efficiency at the spallation position is within 6%for 80%of the samples,with an overall average error within 10%.It is concluded from the rapid prediction model that when the depth of coating spallation increases,the closer the spallation position is to the leading edge of the blade,the greater the difference in cooling efficiency is,and the degree of influence of coating spallation on the cooling efficiency also increases.
