针对密度峰值聚类算法DPC(clustering by fast search and find of density peaks)时间复杂度高、准确度低的缺陷,提出了一种基于Ball-Tree优化的快速密度峰值聚类算法BT-DPC。算法利用第k近邻度量样本局部密度,通过构建Ball-Tree加速...针对密度峰值聚类算法DPC(clustering by fast search and find of density peaks)时间复杂度高、准确度低的缺陷,提出了一种基于Ball-Tree优化的快速密度峰值聚类算法BT-DPC。算法利用第k近邻度量样本局部密度,通过构建Ball-Tree加速密度ρ及距离δ的计算;在类簇分配阶段,结合k近邻思想设计统计学习分配策略,将边界点正确归类。通过在UCI数据集上的实验,将该算法与原密度峰值聚类算法及其改进算法进行了对比,实验结果表明,BT-DPC算法在降低时间复杂度的同时提高了聚类的准确度。展开更多
Using a triangular lattice model to study the designability of proteinfolding, we overcame the parity problem of previous cubic lattice model and enumerated all thesequences and compact structures on a simple two-dime...Using a triangular lattice model to study the designability of proteinfolding, we overcame the parity problem of previous cubic lattice model and enumerated all thesequences and compact structures on a simple two-dimensional triangular lattice model of size4+5+6+5+4. We used two types of amino acids, hydrophobic and polar, to make up the sequences, andachieved 2^(23)+2^(12) different sequences excluding the reverse symmetry sequences. The totalstring number of distinct compact structures was 219,093, excluding reflection symmetry in theself-avoiding path of length 24 triangular lattice model. Based on this model, we applied a fastsearch algorithm by constructing a cluster tree. The algorithm decreased the computation bycomputing the objective energy of non-leaf nodes. The parallel experiments proved that the fast treesearch algorithm yielded an exponential speed-up in the model of size 4+5+6+5+4. Designabilityanalysis was performed to understand the search result.展开更多
为提高果园机器人在果园中作业的自主性、安全性和效率,需要进行有效合理的运动规划。针对传统RRT^(*)(Rapidly exploring random tree star)全局路径规划算法在连续走廊式环境下存在搜索效率低、采样点利用率低、生成路径折线多转角大...为提高果园机器人在果园中作业的自主性、安全性和效率,需要进行有效合理的运动规划。针对传统RRT^(*)(Rapidly exploring random tree star)全局路径规划算法在连续走廊式环境下存在搜索效率低、采样点利用率低、生成路径折线多转角大等问题,以阿克曼底盘果园喷雾机器人为运动模型,提出一种改进双向RRT^(*)的果园喷雾机器人运动规划算法。首先,根据激光雷达建立果园二维平面地图,将果树和障碍物均视为障碍物区域,并结合喷雾机器人本体尺寸,对障碍物进行膨胀化处理;然后,通过改进双向RRT^(*)算法搜索路径,搜索路径过程中结合动态末梢节点导向和势场导向进行偏置采样,并对初步生成的路径进行路径点去冗余以及相邻折线段转角约束处理;最后,采用三阶准均匀B样条曲线对处理后的路径点进行轨迹优化,在优化过程中主要考虑轨迹的碰撞检测和喷雾机器人底盘曲率约束。试验结果表明,相较于传统双向RRT^(*)算法,本文所提出的改进算法规划时间平均减少57.5%,采样点利用率平均提高28.55个百分点,最终路径长度平均缩短7.14%;经三阶准均匀B样条曲线优化后所得轨迹在有、无障碍物两种环境下均满足喷雾机器人最大曲率约束,且仅在换行以及障碍物处存在转弯行为,符合喷雾机器人作业轨迹条件,提高了喷雾机器人的工作效率和自主性。展开更多
The description of complex configuration is a difficult issue.We present a powerful technique for cluster identification and characterization.The scheme is designed to treat and analyze the experimental and/or simulat...The description of complex configuration is a difficult issue.We present a powerful technique for cluster identification and characterization.The scheme is designed to treat and analyze the experimental and/or simulation data from various methods.The main steps are as follows.We first divide the space using face or volume elements from discrete points.Then,we combine the elements with the same and/or similar properties to construct clusters with special physical characterizations.In the algorithm,we adopt an administrative structure of a hierarchy-tree for spatial bodies such as points,lines,faces,blocks,and clusters.Two fast search algorithms with the complexity lnN are generated.The establishment of the hierarchy-tree and the fast searching of spatial bodies are general,which are independent of spatial dimensions.Therefore,it is easy to extend the method to other fields.As a verification and validation,we applied this method and analyzed some two-dimensional and three-dimensional random data.展开更多
文摘针对密度峰值聚类算法DPC(clustering by fast search and find of density peaks)时间复杂度高、准确度低的缺陷,提出了一种基于Ball-Tree优化的快速密度峰值聚类算法BT-DPC。算法利用第k近邻度量样本局部密度,通过构建Ball-Tree加速密度ρ及距离δ的计算;在类簇分配阶段,结合k近邻思想设计统计学习分配策略,将边界点正确归类。通过在UCI数据集上的实验,将该算法与原密度峰值聚类算法及其改进算法进行了对比,实验结果表明,BT-DPC算法在降低时间复杂度的同时提高了聚类的准确度。
文摘Using a triangular lattice model to study the designability of proteinfolding, we overcame the parity problem of previous cubic lattice model and enumerated all thesequences and compact structures on a simple two-dimensional triangular lattice model of size4+5+6+5+4. We used two types of amino acids, hydrophobic and polar, to make up the sequences, andachieved 2^(23)+2^(12) different sequences excluding the reverse symmetry sequences. The totalstring number of distinct compact structures was 219,093, excluding reflection symmetry in theself-avoiding path of length 24 triangular lattice model. Based on this model, we applied a fastsearch algorithm by constructing a cluster tree. The algorithm decreased the computation bycomputing the objective energy of non-leaf nodes. The parallel experiments proved that the fast treesearch algorithm yielded an exponential speed-up in the model of size 4+5+6+5+4. Designabilityanalysis was performed to understand the search result.
文摘为提高果园机器人在果园中作业的自主性、安全性和效率,需要进行有效合理的运动规划。针对传统RRT^(*)(Rapidly exploring random tree star)全局路径规划算法在连续走廊式环境下存在搜索效率低、采样点利用率低、生成路径折线多转角大等问题,以阿克曼底盘果园喷雾机器人为运动模型,提出一种改进双向RRT^(*)的果园喷雾机器人运动规划算法。首先,根据激光雷达建立果园二维平面地图,将果树和障碍物均视为障碍物区域,并结合喷雾机器人本体尺寸,对障碍物进行膨胀化处理;然后,通过改进双向RRT^(*)算法搜索路径,搜索路径过程中结合动态末梢节点导向和势场导向进行偏置采样,并对初步生成的路径进行路径点去冗余以及相邻折线段转角约束处理;最后,采用三阶准均匀B样条曲线对处理后的路径点进行轨迹优化,在优化过程中主要考虑轨迹的碰撞检测和喷雾机器人底盘曲率约束。试验结果表明,相较于传统双向RRT^(*)算法,本文所提出的改进算法规划时间平均减少57.5%,采样点利用率平均提高28.55个百分点,最终路径长度平均缩短7.14%;经三阶准均匀B样条曲线优化后所得轨迹在有、无障碍物两种环境下均满足喷雾机器人最大曲率约束,且仅在换行以及障碍物处存在转弯行为,符合喷雾机器人作业轨迹条件,提高了喷雾机器人的工作效率和自主性。
基金supported by the National Natural Science Foundation of China (Grant Nos 10702010 and 10775018)the Science Foundations of the Laboratory of Computational Physics and China Academy of Engineering Physics (Grant Nos.2009A0102005 and 2009B0101012)
文摘The description of complex configuration is a difficult issue.We present a powerful technique for cluster identification and characterization.The scheme is designed to treat and analyze the experimental and/or simulation data from various methods.The main steps are as follows.We first divide the space using face or volume elements from discrete points.Then,we combine the elements with the same and/or similar properties to construct clusters with special physical characterizations.In the algorithm,we adopt an administrative structure of a hierarchy-tree for spatial bodies such as points,lines,faces,blocks,and clusters.Two fast search algorithms with the complexity lnN are generated.The establishment of the hierarchy-tree and the fast searching of spatial bodies are general,which are independent of spatial dimensions.Therefore,it is easy to extend the method to other fields.As a verification and validation,we applied this method and analyzed some two-dimensional and three-dimensional random data.
