一种改进的行人导航算法研究

发布时间：2018-05-07 08:19

  本文选题：行人导航定位 + 神经网络(BP)　； 参考：《重庆邮电大学学报(自然科学版)》2016年02期

【摘要】：目前大多行人导航系统的研究都以手持手机、绑在腰间等单一姿态为前提,并未考虑行人使用手机过程中同时存在接电话、放在上衣口袋等多种姿态的情况。分析了行人使用手机的4种常见姿态,基于行人航迹推算(pedestrian dead reckoning,PDR)算法设计了一种改进的行人导航算法。利用神经网络(back propagation,BP)模型对手机姿态进行智能识别,设计了一种新的步态检测算法来对多姿态下的步态进行有效检测,并将行人行走过程分为单一姿态与姿态切换过程,对2个过程的航向进行了修正。软件测试结果表明,多姿态下实现系统定位精度在20‰以内,达标率在85%以上,应用于导航系统能够提供准确、可靠、持续的位置信息。
[Abstract]:At present, most of the research on pedestrian navigation system is based on handheld mobile phone, tied to the waist and other single posture, and does not consider that there are many gestures such as answering the phone and putting it in the coat pocket in the process of using the mobile phone. Based on the analysis of four common postures of mobile phones, an improved pedestrian navigation algorithm is designed based on the pedestrian trajectory estimation algorithm. A new gait detection algorithm is designed based on the neural network back propagation (BP) model to detect the mobile phone posture effectively, and the walking process is divided into a single posture and attitude switching process. The course of the two processes is modified. The software test results show that the positioning accuracy of the system is less than 20 鈥，

本文编号：1856162