会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 1. 发明专利
    • Estimation device
    • 估计装置
    • JP2014169923A
    • 2014-09-18
    • JP2013041882
    • 2013-03-04
    • Denso Corp株式会社デンソーDenso It Laboratory Inc株式会社デンソーアイティーラボラトリ
    • SAKAMOTO MAISUZUKI KOICHIRO
    • G01S13/66G01S7/02G01S7/32G01S13/38
    • G01B21/16G01S11/02G01S13/347G01S13/38G01S13/42G01S13/584G01S13/726
    • PROBLEM TO BE SOLVED: To highly accurately estimate a distance to a target on the basis of an observation value.SOLUTION: A control unit is configured to: correct an observation value (Rand V) in a predetermined period of time about a distance R of a target obtained from a dual frequency CW method radar device and a velocity V by a simple linear Kalman filter, calculate a time integration value of the observation value Vuntil corresponding time since start time of the period of time as an observation value δof an amount of displacement δ for each time on the basis of an observation value Vof the velocity V of the post-corrected observation value (Rand V) (S310); and further utilize that a relation between the distance R and the amount of displacement δ is expressed by a formula R=R0+δ using a distance R0 of the target when the amount of displacement δ=0, adopt the formula into a regression equation, execute a regression analysis using the observation value Rand the observation value δas a sample, and estimate that the distance R0 calculated by the regression analysis when the amount of displacement δ=0 is a distance R to the target at the start time (S330 and S340).
    • 要解决的问题:基于观测值高精度地估计到目标的距离。解决方案:控制单元被配置为:在预定时间段内修正观测值(Rand V) 从双频CW方法雷达装置得到的目标和通过简单线性卡尔曼滤波器的速度V计算从时间段起始时间起的观测值Vuntil对应时间的时间积分值作为观测值δ 基于修正后观测值(Rand V)的速度V的观测值V,每次的位移δ(S310); 并且进一步利用当位移量δ= 0时,使用目标的距离R0,通过公式R = R0 +δ来表示距离R和位移量δ之间的关系,将公式转换为回归方程, 使用观测值将观测值Rand表示为样本的观测值δ,并且估计当位移量δ= 0时与回归分析计算出的距离R0是在开始时刻与目标的距离R(S330和S340 )。
    • 2. 发明专利
    • Radar device
    • 雷达设备
    • JP2010071958A
    • 2010-04-02
    • JP2008243145
    • 2008-09-22
    • Denso CorpDenso It Laboratory Inc株式会社デンソー株式会社デンソーアイティーラボラトリ
    • SAKAMOTO MAINATSUME KAZUMAWATANABE MASARUSHIRAI HIDEKIYAMANO CHIHARU
    • G01S13/34G01S7/02
    • G01S13/345G01S3/74G01S13/42G01S13/931G01S2013/9375
    • PROBLEM TO BE SOLVED: To provide a radar device capable of acquiring target information highly accurately.
      SOLUTION: This radar device of an FMCW system generates snapshot data from a beat signal showing a reception result in each modulation period. Then, each autocorrelation matrix in each modulation period generated from snapshot data in each modulation period is averaged as many as a plurality of period portions, and a target azimuth is determined by MUSIC method based on the averaged autocorrelation matrix. In this case, averaging is performed by a weighted mean method based on the amount (interference amount) of noise components included in the snapshot data in each modulation period. Namely, a weighting coefficient to be applied to the autocorrelation matrix in each modulation period is set at a value corresponding to the interference amount of the modulation period (S570). When the interference amount is small, the weighting coefficient is set high, and when the interference amount is large, the weighting coefficient is set low.
      COPYRIGHT: (C)2010,JPO&INPIT
    • 要解决的问题:提供能够高精度地获取目标信息的雷达装置。 解决方案:FMCW系统的该雷达装置从每个调制周期中显示接收结果的拍频信号产生快照数据。 然后,在每个调制周期中从快照数据生成的每个调制周期中的每个自相关矩阵被平均为多个周期部分,并且通过基于平均自相关矩阵的MUSIC方法来确定目标方位角。 在这种情况下,通过基于每个调制周期中包含在快照数据中的噪声分量的量(干扰量)的加权平均方法进行平均。 也就是说,将在每个调制周期中应用于自相关矩阵的加权系数设置为与调制周期的干扰量量对应的值(S570)。 当干扰量小时,加权系数设定为高,当干扰量大时,将加权系数设定得较低。 版权所有(C)2010,JPO&INPIT
    • 3. 发明专利
    • Electronic scanning type radar device
    • 电子扫描型雷达装置
    • JP2009121826A
    • 2009-06-04
    • JP2007292856
    • 2007-11-12
    • Denso CorpDenso It Laboratory Inc株式会社デンソー株式会社デンソーアイティーラボラトリ
    • SHIRAI HIDEKIYAMANO CHIHARUNATSUME KAZUMAWATANABE MASARUSAKAMOTO MAI
    • G01S13/34
    • G01S7/023G01S7/2813G01S13/345G01S13/931
    • PROBLEM TO BE SOLVED: To provide an electronic scanning type radar device capable of suppressing interference by a simple constitution in the electronic scanning type radar device. SOLUTION: Sampled beat signals RD are cut out as a plurality of short-time data SD in a time direction relative to each antenna element. An interference component frequency of an interference wave is detected from a frequency spectrum of the short-time data SD. A phase correction table of an interference component is generated based on a frequency spectrum component of in the detected interference component frequency. Each phase of interference components of reception data of each channel is aligned based on the phase correction table. An interference direction component is suppressed by removing a direct component from the reception data of each channel having each aligned phase. Each phase of the reception data of each channel from which the interference direction component is removed is returned to the original state. COPYRIGHT: (C)2009,JPO&INPIT
    • 解决的问题:提供能够通过电子扫描型雷达装置中的简单结构来抑制干扰的电子扫描型雷达装置。 解决方案:采样差拍信号RD相对于每个天线元件在时间方向上被切割为多个短时数据SD。 从短时数据SD的频谱检测干扰波的干扰分量频率。 基于检测到的干扰分量频率的频谱分量,生成干扰分量的相位校正表。 每个通道的接收数据的干扰分量的每个相位基于相位校正表来对准。 通过从具有每个对齐相位的每个信道的接收数据中去除直接分量来抑制干扰方向分量。 将干扰方向分量从其去除的每个通道的接收数据的每个相位返回到原始状态。 版权所有(C)2009,JPO&INPIT
    • 5. 发明专利
    • Electronic scanning type radar device
    • 电子扫描型雷达装置
    • JP2007232383A
    • 2007-09-13
    • JP2006050682
    • 2006-02-27
    • Denso CorpDenso It Laboratory Inc株式会社デンソー株式会社デンソーアイティーラボラトリ
    • YAMANO CHIHARUNATSUME KAZUMAWATANABE MASARUSAKAMOTO MAI
    • G01S13/32G01S7/35
    • G01S7/4021G01S7/023G01S7/352G01S13/931G01S2007/356
    • PROBLEM TO BE SOLVED: To provide an electronic scanning type radar device capable of interference suppression utilizing directivity, even in a time-sharing reception type electronic scanning type radar. SOLUTION: A sampled beat signal RD is cut out into a plurality of short-time data SD in the time direction relative to each antenna element. An interference component frequency of an interference wave is detected from a frequency spectrum of the short-time data SD. A plurality of turnaround front frequency candidates (f BB ) of the interference wave are generated from the interference component frequency of the interference wave, and phase correction is performed relative to each candidate. Digital beam forming processing is performed relative to each corrected frequency, and the maximum peak of the power in the azimuth direction is extracted respectively, and a frequency candidate showing the maximum peak power is selected, and the incoming azimuth of the interference component is estimated. A filter for suppressing the interference component is applied from the estimated incoming azimuth of the interference component, to thereby suppress the interference component. COPYRIGHT: (C)2007,JPO&INPIT
    • 要解决的问题:即使在分时接收型电子扫描型雷达中,也提供能够利用指向性进行干扰抑制的电子扫描型雷达装置。 解决方案:采样的差拍信号RD相对于每个天线元件在时间方向被切割成多个短时数据SD。 从短时数据SD的频谱检测干扰波的干扰分量频率。 从干扰波的干扰分量频率产生干扰波的多个周转前频率候选(f BB ),并且相对于每个候选进行相位校正。 对每个校正频率进行数字波束形成处理,分别提取方位角方向上的功率的最大峰值,并且选择表示最大峰值功率的频率候选,并且估计干扰分量的输入方位角。 从干扰分量的估计输入方位角施加用于抑制干扰分量的滤波器,从而抑制干扰分量。 版权所有(C)2007,JPO&INPIT
    • 6. 发明专利
    • Noise floor intensity calculation method and fmcw radar
    • 噪声强度计算方法和FMCW雷达
    • JP2008232833A
    • 2008-10-02
    • JP2007072894
    • 2007-03-20
    • Denso Corp株式会社デンソー
    • SAKAMOTO MAI
    • G01S13/34G01S7/40
    • G01S7/354G01S13/345G01S2007/356
    • PROBLEM TO BE SOLVED: To improve detection accuracy of noise floor intensity which is necessary information for interference countermeasures in an FMCW radar.
      SOLUTION: Frequency analysis of a beat signal is performed in each channel and in each section (a rising section, a lowering section), based on sampling data of over-sampled beat signals (S110-S120). A power value distribution histogram using the intensity (power value) of a frequency component as a scale is created based on the frequency component in a high frequency domain having a higher frequency than a detection frequency domain corresponding to a detection range of a target among analysis results (S130), and a power value at which the frequency becomes maximum in the power value distribution histogram is extracted as the noise floor intensity (S140), and if the extracted power value (noise floor intensity) is higher than an interference determination threshold (S150:YES), it is determined that interference with another radar occurs, and interference countermeasure processing is executed (S200).
      COPYRIGHT: (C)2009,JPO&INPIT
    • 要解决的问题:提高噪声基底强度的检测精度,这是FMCW雷达中干扰对策的必要信息。 解决方案:基于过采样拍频信号的采样数据(S110-S120),在每个通道和每个部分(上升部分,下降部分)中执行拍频信号的频率分析。 基于频率高于对应于分析对象的检测范围的检测频域的检测频域的高频域,生成使用频率分量的强度(功率值)作为标度的功率值分布直方图 提取结果(S130),并且提取功率值分布直方图中的频率变得最大的功率值作为噪声本底强度(S140),并且如果提取的功率值(噪声底层强度)高于干扰判定阈值 (S150:是),确定发生与另一雷达的干扰,执行干扰对策处理(S200)。 版权所有(C)2009,JPO&INPIT
    • 7. 发明专利
    • Interference determination method and fmcw radar
    • 干扰测定方法和FMCW雷达
    • JP2008232832A
    • 2008-10-02
    • JP2007072886
    • 2007-03-20
    • Denso Corp株式会社デンソー
    • SAKAMOTO MAI
    • G01S13/34
    • G01S7/021G01S7/023G01S13/345G01S13/931G01S2013/0263G01S2013/9321G01S2013/9325
    • PROBLEM TO BE SOLVED: To improve accuracy of interference determination with another radar in an FMCW radar.
      SOLUTION: A peak frequency component exceeding an excessive determination value is detected in a high frequency domain having a higher frequency than a detection frequency domain corresponding to a detection range of a target based on a result acquired by performing frequency analysis of sampling data of an over-sampled beat signal (S210). A range of a frequency width fd having a frequency of the detected peak frequency component as a center is set as an excessive domain (S230), and a mean value of signal intensities on a pair of domain edges in the excessive domain is determined as an interpolation value in each set excessive domain, and after replacing signal intensities of all the frequency components belonging to the excessive domain with the interpolation value (S240), each intensity of the frequency components in integrated over the whole high frequency domain (S250), and existence of interference with another radar is determined based on the integrated value.
      COPYRIGHT: (C)2009,JPO&INPIT
    • 要解决的问题:提高FMCW雷达中另一台雷达的干扰确定精度。 解决方案:基于通过执行采样数据的频率分析获得的结果,在具有比与目标的检测范围相对应的检测频域更高的频率的高频域中检测超过过度判定值的峰值频率分量 的过采样拍频信号(S210)。 将检测到的峰值频率分量的频率作为中心的频率宽度fd的范围被设定为过度域(S230),并且将过度域中的一对域边缘上的信号强度的平均值确定为 每个集合过多域中的内插值,并且在用内插值替换属于过度域的所有频率分量的信号强度(S240)之后,在整个高频域上集成的每个频率分量的强度(S250),以及 基于积分值确定与另一雷达的干扰的存在。 版权所有(C)2009,JPO&INPIT
    • 8. 发明专利
    • Vehicular road shape recognition method and device, and recording medium
    • 车道形状识别方法和装置以及记录介质
    • JP2012242937A
    • 2012-12-10
    • JP2011110224
    • 2011-05-17
    • Denso Corp株式会社デンソー
    • SAGAWA YOSHIESAKAMOTO MAINOZAWA TOYOJI
    • G08G1/16B60R21/00G01S13/93
    • PROBLEM TO BE SOLVED: To provide a vehicular road shape recognition method and device capable of frequently calculating a more accurate road shape, and a recording medium.SOLUTION: The vehicular road shape recognition method includes: determining the recognition classification as to whether an object is a moving object or a stationary object on the basis of relative speed of the object and self-vehicle speed; extracting object unit data effective to recognize a road shape on the basis of a determination result of the recognition classification; extracting a horizontal position of a stationary object with the shortest distance, in the vehicle width direction, to the self-vehicle side from among the extracted object unit data and determining a stationary object, which is positioned within a range apart from the horizontal position of the stationary object closest to the self-vehicle in the vehicle width direction by a predetermined distance and has the shortest direct distance from the self-vehicle, as an original point; then forming data on a road side object group by connecting data pieces where the distance monotonously increases as a connection condition from the original point and grouping them; and recognizing a road edge on the basis of the formed data on the road side object group.
    • 要解决的问题:提供能够频繁地计算更准确的道路形状的车辆道路形状识别方法和装置以及记录介质。 解决方案:车辆道路形状识别方法包括:基于物体的相对速度和自车速度确定对象是移动物体还是静止物体的识别分类; 基于所述识别分类的确定结果提取有效地识别道路形状的对象单位数据; 从所提取的对象单元数据中,将车辆宽度方向上的距离最短的静止物体的水平位置提取到自车侧,并且确定位于距离所提取的对象单位数据的水平位置以外的范围内的静止物体 作为原始点,在车辆宽度方向上最接近自车的固定物体预定距离并且具有与自车的最短直接距离; 然后通过将距离单调增加的数据作为连接状态从原点连接并分组,形成路侧对象组的数据; 并基于道路侧对象组上形成的数据来识别路边。 版权所有(C)2013,JPO&INPIT
    • 9. 发明专利
    • Vehicular road shape recognition method and device, and recording medium
    • 车道形状识别方法和装置以及记录介质
    • JP2012242934A
    • 2012-12-10
    • JP2011110221
    • 2011-05-17
    • Denso Corp株式会社デンソー
    • SAGAWA YOSHIESAKAMOTO MAINOZAWA TOYOJI
    • G08G1/16G01S13/93
    • PROBLEM TO BE SOLVED: To provide a vehicular road shape recognition method and device capable of frequently calculating a more accurate road shape, and a recording medium.SOLUTION: The vehicular road shape recognition method includes: determining the recognition classification as to whether an object is a moving object or a stationary object on the basis of relative speed of the object and self-vehicle speed; extracting object unit data effective to recognize a road shape on the basis of a determination result of the recognition classification; forming data on a road side object group on the basis of the object unit data; and recognizing the road edge on the basis of the formed data on a road side object group. The series of processing are repeatedly executed at a predetermined cycle. After the extraction processing, data addition processing is performed for adding the object unit data obtained by the extraction processing in a previous cycle to the object unit data obtained by the extraction processing in a present cycle. In the recognition processing, the road edge is recognized based on the object unit data obtained by the data addition processing.
    • 要解决的问题:提供能够频繁地计算更准确的道路形状的车辆道路形状识别方法和装置以及记录介质。 解决方案:车辆道路形状识别方法包括:基于物体的相对速度和自车速度确定对象是移动物体还是静止物体的识别分类; 基于所述识别分类的确定结果提取有效地识别道路形状的对象单位数据; 基于对象单位数据在道路侧对象组上形成数据; 并且基于路侧对象组上形成的数据来识别路边。 一系列处理以预定的周期重复执行。 在提取处理之后,执行数据相加处理,用于将通过前一周期中的提取处理获得的对象单元数据添加到在当前周期中通过提取处理获得的对象单元数据。 在识别处理中,基于通过数据附加处理获得的对象单位数据来识别路边。 版权所有(C)2013,JPO&INPIT
    • 10. 发明专利
    • Recognition method and device of road shape for vehicle, and recording medium
    • 车辆道路形状的识别方法和装置以及记录介质
    • JP2012198730A
    • 2012-10-18
    • JP2011061860
    • 2011-03-21
    • Denso Corp株式会社デンソー
    • SAGAWA YOSHIENOZAWA TOYOJISAKAMOTO MAIOKATA KOJI
    • G08G1/16G01S13/89G01S17/89
    • PROBLEM TO BE SOLVED: To provide a recognition method and device of a road shape for a vehicle and a recording medium, which can frequently calculate a further exact road shape in a demerit scene.SOLUTION: Based on a distance to an object and an angle in the vehicle width direction, the road shape of a vehicle front is recognized, and based on the recognized road shape and the recognition degree, a possibility of the object staying on the same lane with an own vehicle is determined, and based on the determination result, a correction value for correcting an own lane probability is calculated. The determination is made whether a deviation exists or not in a curvature of the own vehicle traveling road and a curvature of the recognized road shape. When the deviation does not exist, the own lane probability is corrected by the correction value, and based on the own lane probability after the correction, a preceding vehicle is selected, while when the deviation exists, the preceding vehicle is selected based on the own lane probability without correcting the own lane probability by the correction value.
    • 要解决的问题:提供一种用于车辆和记录介质的道路形状的识别方法和装置,其可以经常在划痕场景中计算更准确的道路形状。 解决方案:基于与物体的距离和车辆宽度方向上的角度,识别车辆前方的道路形状,并且基于所识别的道路形状和识别度,使物体停留的可能性 确定与自身车辆相同的车道,并且基于确定结果,计算用于校正自身车道概率的校正值。 确定在本车辆行驶道路的曲率中是否存在偏差以及公认道路形状的曲率。 当偏差不存在时,通过校正值校正自身车道概率,并且基于校正后的自身车道概率,选择前一车辆,而当存在偏差时,基于自己的车辆选择前一车辆 车道概率,而不用校正值校正自己的车道概率。 版权所有(C)2013,JPO&INPIT