针对目前原始自适应蒙特卡洛定位(Adaptive Monte Carlo Localization,AMCL)在相似环境下绑架检测容易出错且重定位极易失败等问题,提出基于墙角族语义尺寸链的改进AMCL算法.融合机器人多传感器信息和Gmapping算法构建二维栅格地图,基于...针对目前原始自适应蒙特卡洛定位(Adaptive Monte Carlo Localization,AMCL)在相似环境下绑架检测容易出错且重定位极易失败等问题,提出基于墙角族语义尺寸链的改进AMCL算法.融合机器人多传感器信息和Gmapping算法构建二维栅格地图,基于Yolov5获取室内环境的目标检测框和类别信息,结合GrabCut算法和贝叶斯方法构建增量式语义映射地图;通过墙角的凸、凹和墙角相对于机器人的方位角对墙角进行分类,充分发掘语义映射地图中各墙角之间、墙角与室内物体之间的类别和位置关系,构建墙角族语义尺寸链和相应检索表;在定位过程中,基于墙角族语义尺寸链进行全局预定位,提出绑架检测机制进行绑架检测,在检测到绑架事件发生后,基于改进AMCL算法实现定位自恢复.最后,通过真实环境下的绑架实验验证了本文方法的有效性,实验表明,所提方法的全局定位准确率、全局定位速率、绑架检测准确率和绑架后定位准确率在相似环境下分别提升了42%、214%、88%和72%;在非相似环境下分别提升了44%、152%、12%和92%;在长走廊环境下分别提升了36%、426%、26%和68%.展开更多
机器人定位技术作为智能机器人领域的重要技术,是机器人进行自主规划和导航的重要前提。为解决机器人运动过程中的绑架问题,在蒙特卡罗定位(Monte Carlo localization, MCL)算法的基础上,提出了基于激光雷达似然域模型的定位可靠度评判...机器人定位技术作为智能机器人领域的重要技术,是机器人进行自主规划和导航的重要前提。为解决机器人运动过程中的绑架问题,在蒙特卡罗定位(Monte Carlo localization, MCL)算法的基础上,提出了基于激光雷达似然域模型的定位可靠度评判算法以及基于惯性导航单元的定位自恢复模型。定位可靠度评判算法对机器人是否发生绑架问题进行判定,当发生绑架问题后,首先基于惯性导航单元的测量数据进行位姿预估计,然后基于预估计的位姿构建粒子重分布模型,最后进行粒子滤波得到重定位的结果,达到了对机器人绑架判定和自恢复定位的目的。经过实验测试和对比,该算法可以对绑架问题进行高效的判断,具有更高的恢复效率和准确度。展开更多
[Objective] This study was to investigate the restoring ability of normal indica red rice Ruby and to carry out its restoring gene mapping. [Method] Normal indica red rice Ruby was hybridized with the sterile lines Zh...[Objective] This study was to investigate the restoring ability of normal indica red rice Ruby and to carry out its restoring gene mapping. [Method] Normal indica red rice Ruby was hybridized with the sterile lines Zhenxian 97A, D62A, G46A and D702A to prepare their F1, BC1 and F2 progenies, and the pollen fertilities of these progenies were investigated. Meanwhile the restoring genes were mapped using SSLP. [ Result] For the sterile lines tested, Ruby has a gene to restore their fertilities. This gene is located on the chromosome 7 and shows a genetic distance of 7.4 cM with RM182. Unlike the clustering distribution of the restoring genes on chromosome 10, it is a specific restoring gene. [ Conclusion] it is feasible to breed restoring genes controlling red color characters via transgene and backcross.展开更多
Deployment of nodes based on K-barrier coverage in an underground wireless sensor network is described. The network has automatic routing recovery by using a basic information table (BIT) for each node. An RSSI positi...Deployment of nodes based on K-barrier coverage in an underground wireless sensor network is described. The network has automatic routing recovery by using a basic information table (BIT) for each node. An RSSI positioning algorithm based on a path loss model in the coal mine is used to calculate the path loss in real time within the actual lane way environment. Simulation results show that the packet loss can be controlled to less than 15% by the routing recovery algorithm under special recovery circum- stances. The location precision is within 5 m, which greatly enhances performance compared to tradi- tional frequency location systems. This approach can meet the needs for accurate location underground.展开更多
文摘针对目前原始自适应蒙特卡洛定位(Adaptive Monte Carlo Localization,AMCL)在相似环境下绑架检测容易出错且重定位极易失败等问题,提出基于墙角族语义尺寸链的改进AMCL算法.融合机器人多传感器信息和Gmapping算法构建二维栅格地图,基于Yolov5获取室内环境的目标检测框和类别信息,结合GrabCut算法和贝叶斯方法构建增量式语义映射地图;通过墙角的凸、凹和墙角相对于机器人的方位角对墙角进行分类,充分发掘语义映射地图中各墙角之间、墙角与室内物体之间的类别和位置关系,构建墙角族语义尺寸链和相应检索表;在定位过程中,基于墙角族语义尺寸链进行全局预定位,提出绑架检测机制进行绑架检测,在检测到绑架事件发生后,基于改进AMCL算法实现定位自恢复.最后,通过真实环境下的绑架实验验证了本文方法的有效性,实验表明,所提方法的全局定位准确率、全局定位速率、绑架检测准确率和绑架后定位准确率在相似环境下分别提升了42%、214%、88%和72%;在非相似环境下分别提升了44%、152%、12%和92%;在长走廊环境下分别提升了36%、426%、26%和68%.
文摘机器人定位技术作为智能机器人领域的重要技术,是机器人进行自主规划和导航的重要前提。为解决机器人运动过程中的绑架问题,在蒙特卡罗定位(Monte Carlo localization, MCL)算法的基础上,提出了基于激光雷达似然域模型的定位可靠度评判算法以及基于惯性导航单元的定位自恢复模型。定位可靠度评判算法对机器人是否发生绑架问题进行判定,当发生绑架问题后,首先基于惯性导航单元的测量数据进行位姿预估计,然后基于预估计的位姿构建粒子重分布模型,最后进行粒子滤波得到重定位的结果,达到了对机器人绑架判定和自恢复定位的目的。经过实验测试和对比,该算法可以对绑架问题进行高效的判断,具有更高的恢复效率和准确度。
基金Supported by Sci-tech Program for Excellent Young Scientists of Sichuan Province(01ZQ052)~~
文摘[Objective] This study was to investigate the restoring ability of normal indica red rice Ruby and to carry out its restoring gene mapping. [Method] Normal indica red rice Ruby was hybridized with the sterile lines Zhenxian 97A, D62A, G46A and D702A to prepare their F1, BC1 and F2 progenies, and the pollen fertilities of these progenies were investigated. Meanwhile the restoring genes were mapped using SSLP. [ Result] For the sterile lines tested, Ruby has a gene to restore their fertilities. This gene is located on the chromosome 7 and shows a genetic distance of 7.4 cM with RM182. Unlike the clustering distribution of the restoring genes on chromosome 10, it is a specific restoring gene. [ Conclusion] it is feasible to breed restoring genes controlling red color characters via transgene and backcross.
基金supported by the National Key Technology R&D Program of China (No. 2008BAH37B05095)
文摘Deployment of nodes based on K-barrier coverage in an underground wireless sensor network is described. The network has automatic routing recovery by using a basic information table (BIT) for each node. An RSSI positioning algorithm based on a path loss model in the coal mine is used to calculate the path loss in real time within the actual lane way environment. Simulation results show that the packet loss can be controlled to less than 15% by the routing recovery algorithm under special recovery circum- stances. The location precision is within 5 m, which greatly enhances performance compared to tradi- tional frequency location systems. This approach can meet the needs for accurate location underground.
