基于KNN运动模式识别的改进PDR室内定位
|
摘要
为提高行人航迹推算(PDR)的定位精度,首先利用K最近邻法(KNN)对智能手机采集的6种行人运动模式数据进行识别,再与基于支持向量机(SVM)和高斯朴素贝叶斯(GNB)的运动模式识别方法进行对比,最后在实际环境下进行运动模式辅助的PDR实验。结果表明,KNN方法不仅比SVM和GNB方法易于实现,而且具有更高的识别正确率。在识别行人运动模式的前提下,PDR的室内定位效果比传统PDR方法定位效果更好。
At first, we used the K nearest neighbor(KNN) method to identify the data of 6 pedestrian motion modes collected by smartphone. And then, we compared this method with the motion mode recognition methods based on support vector machine(SVM) and Gaussian naive Bayes(GNB). Finally, we carried out pedestrian dead reckoning(PDR) experiment based on pedestrian motion mode in the actual environment. The results show that the KNN method is not only easy to realize, but also has higher recognition accuracy compared with SVM and GNB methods. The location result of PDR is excellent than the traditional PDR method on the premise of identifying the pedestrian motion mode.
At first, we used the K nearest neighbor(KNN) method to identify the data of 6 pedestrian motion modes collected by smartphone. And then, we compared this method with the motion mode recognition methods based on support vector machine(SVM) and Gaussian naive Bayes(GNB). Finally, we carried out pedestrian dead reckoning(PDR) experiment based on pedestrian motion mode in the actual environment. The results show that the KNN method is not only easy to realize, but also has higher recognition accuracy compared with SVM and GNB methods. The location result of PDR is excellent than the traditional PDR method on the premise of identifying the pedestrian motion mode.
引文
[1]Aminian K,Robert P,Buchser E,et al.Physical Activity Monitoring Based on Accelerometry:Validation and Comparison with Video Observation[J].Medical and Biological Engineering and Computing,1999,37(3):304-308
[2]Lara O D,Labrador M A.A Survey on Human Activity Recognition Using Wearable Sensors[J].IEEE Communications Surveys and Tutorials,2013,15(3):1 192-1 209
[3]Tharwat A,Mahdi H,Elhoseny M,et al.Recognizing Human Activity in Mobile Crowdsensing Environment Using Optimized KNN Algorithm[J].Expert Systems with Applications,2018(4):507-523
[4]Elhoushi M,Georgy J,Noureldin A,et al.Motion Mode Recognition for Indoor Pedestrian Navigation Using Portable Devices[J].IEEE Transactions on Instrumentation and Measurement,2015,65(1):208-221
[5]赵展,夏旺,闫利.基于特征距离的多类SVM分类方法研究[J].地理空间信息,2017,15(11):84-87
[6]DAI W Y,XUE G R,YANG Q,et al.Transferring Naive Bayes Classifiers for Text Classification[C].National Conference on Artificial Intelligence,AAAI Press,2007:540-545
[7]陈振洲,李磊,姚正安.基于SVM的特征加权KNN算法[J].中山大学学报(自然科学版),2005,44(1):17-20
[8]奉国和,吴敬学.KNN分类算法改进研究进展[J].图书情报工作,2012,56(21):97-100
[9]郭际明,王威,张帅.支持向量机辅助下的智能手机行人运动模式识别[J].测绘通报,2018(2):1-5
[10]Varma S,Simon R.Bias in Error Estimation When Using Crossvalidation for Model Selection[J].Bmc Bioinformatics,2006,7(1):1-8
[11]LIN F,Cohen W W.Semi-supervised Classification of Network Data Using Very Few Labels[C].IEEE International Conference on Advances in Social Networks Analysis and Mining,2010:192-199
[12]陈国良,张言哲,汪云甲,等.WiFi-PDR室内组合定位的无迹卡尔曼滤波算法[J].测绘学报,2015,44(12):1 314-1 321
[2]Lara O D,Labrador M A.A Survey on Human Activity Recognition Using Wearable Sensors[J].IEEE Communications Surveys and Tutorials,2013,15(3):1 192-1 209
[3]Tharwat A,Mahdi H,Elhoseny M,et al.Recognizing Human Activity in Mobile Crowdsensing Environment Using Optimized KNN Algorithm[J].Expert Systems with Applications,2018(4):507-523
[4]Elhoushi M,Georgy J,Noureldin A,et al.Motion Mode Recognition for Indoor Pedestrian Navigation Using Portable Devices[J].IEEE Transactions on Instrumentation and Measurement,2015,65(1):208-221
[5]赵展,夏旺,闫利.基于特征距离的多类SVM分类方法研究[J].地理空间信息,2017,15(11):84-87
[6]DAI W Y,XUE G R,YANG Q,et al.Transferring Naive Bayes Classifiers for Text Classification[C].National Conference on Artificial Intelligence,AAAI Press,2007:540-545
[7]陈振洲,李磊,姚正安.基于SVM的特征加权KNN算法[J].中山大学学报(自然科学版),2005,44(1):17-20
[8]奉国和,吴敬学.KNN分类算法改进研究进展[J].图书情报工作,2012,56(21):97-100
[9]郭际明,王威,张帅.支持向量机辅助下的智能手机行人运动模式识别[J].测绘通报,2018(2):1-5
[10]Varma S,Simon R.Bias in Error Estimation When Using Crossvalidation for Model Selection[J].Bmc Bioinformatics,2006,7(1):1-8
[11]LIN F,Cohen W W.Semi-supervised Classification of Network Data Using Very Few Labels[C].IEEE International Conference on Advances in Social Networks Analysis and Mining,2010:192-199
[12]陈国良,张言哲,汪云甲,等.WiFi-PDR室内组合定位的无迹卡尔曼滤波算法[J].测绘学报,2015,44(12):1 314-1 321