Since the position of the electron in a hydrogen atom cannot be determined, the region in which it resides is said to be determined stochastically and forms an electron cloud. The probability density function of the s...Since the position of the electron in a hydrogen atom cannot be determined, the region in which it resides is said to be determined stochastically and forms an electron cloud. The probability density function of the single electron in 1s orbit is expressed as φ2, a function of distance from the nucleus. However, the probability of existence of the electron is expressed as a radial distribution function at an arbitrary distance from the nucleus, so it is estimated as the probability of the entire spherical shape of that radius. In this study, it has been found that the electron existence probability approximates the radial distribution function by assuming that the probability of existence of the electron being in the vicinity of the nucleus follows a normal distribution for arbitrary x-, y-, and z-axis directions. This implies that the probability of existence of the electron, which has been known only from the distance information, would follow a normal distribution independently in the three directions. When the electrons’ motion is extremely restricted in a certain direction by the magnetic field of both tokamak and helical fusion reactors, the probability of existence of the electron increases with proximity to the nucleus, and as a result, it is less likely to be liberated from the nucleus. Therefore, more and more energy is required to free the nucleus from the electron in order to generate plasma.展开更多
The autonomous exploration and mapping of an unknown environment is useful in a wide range of applications and thus holds great significance. Existing methods mostly use range sensors to generate twodimensional (2D) g...The autonomous exploration and mapping of an unknown environment is useful in a wide range of applications and thus holds great significance. Existing methods mostly use range sensors to generate twodimensional (2D) grid maps. Red/green/blue-depth (RGB-D) sensors provide both color and depth information on the environment, thereby enabling the generation of a three-dimensional (3D) point cloud map that is intuitive for human perception. In this paper, we present a systematic approach with dual RGB-D sensors to achieve the autonomous exploration and mapping of an unknown indoor environment. With the synchronized and processed RGB-D data, location points were generated and a 3D point cloud map and 2D grid map were incrementally built. Next, the exploration was modeled as a partially observable Markov decision process. Partial map simulation and global frontier search methods were combined for autonomous exploration, and dynamic action constraints were utilized in motion control. In this way, the local optimum can be avoided and the exploration efficacy can be ensured. Experiments with single connected and multi-branched regions demonstrated the high robustness, efficiency, and superiority of the developed system and methods.展开更多
Based on the fast algorithm of meteorological satellite guide wind vector tracing, cloud motion wind vector is calculated. According to the different characteristics of cloud motion wind field and sounding wind field,...Based on the fast algorithm of meteorological satellite guide wind vector tracing, cloud motion wind vector is calculated. According to the different characteristics of cloud motion wind field and sounding wind field, a method which fuses conventional data with unconventional data based on variation principle is presented. The fundamental is constructing a cost function that makes the value approach conventional data and the gradient approach unconventional data. Using this method, the conventional wind and the cloud motion wind are fused. The fused wind field has high resolu- tion. Its wind direction approaches cloud motion wind which indicates move direction of the synoptic system, and its velocity approaches conventional wind which indicates move velocity of the synoptic system. The wind field data are used for short-time forecast of severe convective weather location, which gets a good result.展开更多
真实场景点云不仅具有点云的空间几何信息,还具有三维物体的颜色信息,现有的网络无法有效利用真实场景的局部特征以及空间几何特征信息,因此提出了一种双通道特征融合的真实场景点云语义分割方法DCFNet(dual-channel feature fusion of ...真实场景点云不仅具有点云的空间几何信息,还具有三维物体的颜色信息,现有的网络无法有效利用真实场景的局部特征以及空间几何特征信息,因此提出了一种双通道特征融合的真实场景点云语义分割方法DCFNet(dual-channel feature fusion of real scene for point cloud semantic segmentation)可用于不同场景下的室内外场景语义分割。更具体地说,为了解决不能充分提取真实场景点云颜色信息的问题,该方法采用上下两个输入通道,通道均采用相同的特征提取网络结构,其中上通道的输入是完整RGB颜色和点云坐标信息,该通道主要关注于复杂物体对象场景特征,下通道仅输入点云坐标信息,该通道主要关注于点云的空间几何特征;在每个通道中为了更好地提取局部与全局信息,改善网络性能,引入了层间融合模块和Transformer通道特征扩充模块;同时,针对现有的三维点云语义分割方法缺乏关注局部特征与全局特征的联系,导致对复杂场景的分割效果不佳的问题,对上下两个通道所提取的特征通过DCFFS(dual-channel feature fusion segmentation)模块进行融合,并对真实场景进行语义分割。对室内复杂场景和大规模室内外场景点云分割基准进行了实验,实验结果表明,提出的DCFNet分割方法在S3DIS Area5室内场景数据集以及STPLS3D室外场景数据集上,平均交并比(MIOU)分别达到71.18%和48.87%,平均准确率(MACC)和整体准确率(OACC)分别达到77.01%与86.91%,实现了真实场景的高精度点云语义分割。展开更多
移动单线激光雷达(Laser detection and ranging,LiDAR)扫描(Mobile single-layer LiDAR scanning,MSLS)树冠叶面积估计方法使用单一视角的单线激光雷达采集树冠点云数据,获取的冠层信息不够全面,限制了树冠叶面积估计精度。本文提出一...移动单线激光雷达(Laser detection and ranging,LiDAR)扫描(Mobile single-layer LiDAR scanning,MSLS)树冠叶面积估计方法使用单一视角的单线激光雷达采集树冠点云数据,获取的冠层信息不够全面,限制了树冠叶面积估计精度。本文提出一种基于移动多线LiDAR扫描(Mobile multi-layer LiDAR scanning,MMLS)的树冠叶面积估计方法,使用多线LiDAR从多个视角采集树冠点云数据,提升树冠叶面积估计精度。首先,将多线LiDAR采集的点云数据变换到世界坐标系下,通过感兴趣区域(Region of interest,ROI)提取出树冠点云。然后,提出一种MMLS树冠点云融合方法,逐个融合单个激光器采集的树冠点云,设置距离阈值删除重复点,添加新点。最后,构建MMLS空间分辨率网格,建立基于树冠网格面积的树冠叶面积估计模型。实验使用VLP-16型多线LiDAR传感器搭建MMLS系统,设置1、1.5 m 2个测量距离和间隔45°的8个测量角度对6个具有不同冠层密度的树冠进行数据采集,共得到96个树冠样本。采用本文方法,树冠叶面积线性估计模型的均方根误差(Root mean squared error,RMSE)为0.1041 m^(2),比MSLS模型降低0.0578 m^(2),决定系数R^(2)为0.9526,比MSLS模型提高0.0675。实验结果表明,本文方法通过多线LiDAR多视角树冠点云数据采集、MMLS树冠点云融合和空间分辨率网格构建,有效提升了树冠叶面积估计精度。展开更多
文摘Since the position of the electron in a hydrogen atom cannot be determined, the region in which it resides is said to be determined stochastically and forms an electron cloud. The probability density function of the single electron in 1s orbit is expressed as φ2, a function of distance from the nucleus. However, the probability of existence of the electron is expressed as a radial distribution function at an arbitrary distance from the nucleus, so it is estimated as the probability of the entire spherical shape of that radius. In this study, it has been found that the electron existence probability approximates the radial distribution function by assuming that the probability of existence of the electron being in the vicinity of the nucleus follows a normal distribution for arbitrary x-, y-, and z-axis directions. This implies that the probability of existence of the electron, which has been known only from the distance information, would follow a normal distribution independently in the three directions. When the electrons’ motion is extremely restricted in a certain direction by the magnetic field of both tokamak and helical fusion reactors, the probability of existence of the electron increases with proximity to the nucleus, and as a result, it is less likely to be liberated from the nucleus. Therefore, more and more energy is required to free the nucleus from the electron in order to generate plasma.
基金the National Natural Science Foundation of China (61720106012 and 61403215)the Foundation of State Key Laboratory of Robotics (2006-003)the Fundamental Research Funds for the Central Universities for the financial support of this work.
文摘The autonomous exploration and mapping of an unknown environment is useful in a wide range of applications and thus holds great significance. Existing methods mostly use range sensors to generate twodimensional (2D) grid maps. Red/green/blue-depth (RGB-D) sensors provide both color and depth information on the environment, thereby enabling the generation of a three-dimensional (3D) point cloud map that is intuitive for human perception. In this paper, we present a systematic approach with dual RGB-D sensors to achieve the autonomous exploration and mapping of an unknown indoor environment. With the synchronized and processed RGB-D data, location points were generated and a 3D point cloud map and 2D grid map were incrementally built. Next, the exploration was modeled as a partially observable Markov decision process. Partial map simulation and global frontier search methods were combined for autonomous exploration, and dynamic action constraints were utilized in motion control. In this way, the local optimum can be avoided and the exploration efficacy can be ensured. Experiments with single connected and multi-branched regions demonstrated the high robustness, efficiency, and superiority of the developed system and methods.
文摘Based on the fast algorithm of meteorological satellite guide wind vector tracing, cloud motion wind vector is calculated. According to the different characteristics of cloud motion wind field and sounding wind field, a method which fuses conventional data with unconventional data based on variation principle is presented. The fundamental is constructing a cost function that makes the value approach conventional data and the gradient approach unconventional data. Using this method, the conventional wind and the cloud motion wind are fused. The fused wind field has high resolu- tion. Its wind direction approaches cloud motion wind which indicates move direction of the synoptic system, and its velocity approaches conventional wind which indicates move velocity of the synoptic system. The wind field data are used for short-time forecast of severe convective weather location, which gets a good result.
文摘真实场景点云不仅具有点云的空间几何信息,还具有三维物体的颜色信息,现有的网络无法有效利用真实场景的局部特征以及空间几何特征信息,因此提出了一种双通道特征融合的真实场景点云语义分割方法DCFNet(dual-channel feature fusion of real scene for point cloud semantic segmentation)可用于不同场景下的室内外场景语义分割。更具体地说,为了解决不能充分提取真实场景点云颜色信息的问题,该方法采用上下两个输入通道,通道均采用相同的特征提取网络结构,其中上通道的输入是完整RGB颜色和点云坐标信息,该通道主要关注于复杂物体对象场景特征,下通道仅输入点云坐标信息,该通道主要关注于点云的空间几何特征;在每个通道中为了更好地提取局部与全局信息,改善网络性能,引入了层间融合模块和Transformer通道特征扩充模块;同时,针对现有的三维点云语义分割方法缺乏关注局部特征与全局特征的联系,导致对复杂场景的分割效果不佳的问题,对上下两个通道所提取的特征通过DCFFS(dual-channel feature fusion segmentation)模块进行融合,并对真实场景进行语义分割。对室内复杂场景和大规模室内外场景点云分割基准进行了实验,实验结果表明,提出的DCFNet分割方法在S3DIS Area5室内场景数据集以及STPLS3D室外场景数据集上,平均交并比(MIOU)分别达到71.18%和48.87%,平均准确率(MACC)和整体准确率(OACC)分别达到77.01%与86.91%,实现了真实场景的高精度点云语义分割。
文摘移动单线激光雷达(Laser detection and ranging,LiDAR)扫描(Mobile single-layer LiDAR scanning,MSLS)树冠叶面积估计方法使用单一视角的单线激光雷达采集树冠点云数据,获取的冠层信息不够全面,限制了树冠叶面积估计精度。本文提出一种基于移动多线LiDAR扫描(Mobile multi-layer LiDAR scanning,MMLS)的树冠叶面积估计方法,使用多线LiDAR从多个视角采集树冠点云数据,提升树冠叶面积估计精度。首先,将多线LiDAR采集的点云数据变换到世界坐标系下,通过感兴趣区域(Region of interest,ROI)提取出树冠点云。然后,提出一种MMLS树冠点云融合方法,逐个融合单个激光器采集的树冠点云,设置距离阈值删除重复点,添加新点。最后,构建MMLS空间分辨率网格,建立基于树冠网格面积的树冠叶面积估计模型。实验使用VLP-16型多线LiDAR传感器搭建MMLS系统,设置1、1.5 m 2个测量距离和间隔45°的8个测量角度对6个具有不同冠层密度的树冠进行数据采集,共得到96个树冠样本。采用本文方法,树冠叶面积线性估计模型的均方根误差(Root mean squared error,RMSE)为0.1041 m^(2),比MSLS模型降低0.0578 m^(2),决定系数R^(2)为0.9526,比MSLS模型提高0.0675。实验结果表明,本文方法通过多线LiDAR多视角树冠点云数据采集、MMLS树冠点云融合和空间分辨率网格构建,有效提升了树冠叶面积估计精度。
