| 双状态机的自适应步态检测算法 |
| |
| 引用本文： | 路泽忠,卢小平,马靓婷,陶晓晓.双状态机的自适应步态检测算法[J].测绘科学,2021,46(1):56-61,75. |
| |
| 作者姓名： | 路泽忠 卢小平 马靓婷 陶晓晓 |
| |
| 作者单位： | 河南理工大学自然资源部矿山时空信息与生态修复重点实验室,河南焦作454003;河南理工大学自然资源部矿山时空信息与生态修复重点实验室,河南焦作454003;河南理工大学自然资源部矿山时空信息与生态修复重点实验室,河南焦作454003;河南理工大学自然资源部矿山时空信息与生态修复重点实验室,河南焦作454003 |
| |
| 基金项目： | 河南省高校创新团队支持计划项目;2016年国家重点研发计划项目;河南省创新型科技创新团队支持计划项目 |
| |
| 摘 要： | 针对步态检测算法无法适应于不同的运动状态、不同传感器姿态问题,该文提出基于双状态机的自适应步态检测算法,该方法在原算法基础上利用最大整体加速度阈值对状态机进行状态分割,并利用时间阈值和整体加速度阈值对状态机进行二次判断。通过四类步态检测算法的比较实验,常规峰值检测算法、状态机步态检测算法在行人行走状态下步态检测精度均优于99%,而在行人跑步状态和行走+跑步状态的步态检测精度仅为60%和70%,无法适应行人的不同运动状态;本文算法在不同行人运动状态、不同传感器姿态下均具有较强的适应性和可靠的步态检测正确率。
|
| 关 键 词： | 行人航迹推算 峰值检测算法 自相关分析法 状态机 最大整体加速度 |
Adaptive gait detection algorithm based on dual state machine |
| |
| LU Zezhong,LU Xiaoping,MA Liangting,TAO Xiaoxiao.Adaptive gait detection algorithm based on dual state machine[J].Science of Surveying and Mapping,2021,46(1):56-61,75. |
| |
| Authors: | LU Zezhong LU Xiaoping MA Liangting TAO Xiaoxiao |
| |
| Institution: | (Key Laboratory of Spatio-temporal Information and Ecological Restoration of Mines,MNR,Henan Polytechnic University,Jiaozuo,Henan 454003,China) |
| |
| Abstract: | To solve the problem of the gait detection algorithm could not adapt to the different motion states and different sensor attitudes,this paper proposed an adaptive gait detection algorithm based on dual state machine.This method used the maximum overall acceleration threshold on the basis of the original algorithm to perform state segmentation,and used the time threshold and the overall acceleration threshold of pedestrians in different motion states to make a second judgment on the state machine.By comparative experiment of four types of gait detection algorithm,the peak detection algorithm and the autocorrelation analysis gait detection algorithm had a detection accuracy of over 99%when pedestrians were in walking state,but a relatively low accuracy of only 60%and 70%when pedestrians were in running state.The algorithm proposed in this paper had strong adaptability and reliable gait detection accuracy in different pedestrian movement states and sensor attitudes. |
| |
| Keywords: | pedestrian dead reckoning peak detection algorithm autocorrelation analysis algorithm state machine maximum overall acceleration |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
