面向非约束智能手机的PDR算法研究
发布时间：2018-11-22 12:38
【摘要】：近几年,室内定位技术在众多的领域都扮演着越来越重要的角色。与此同时,随着微电机系统的大力发展,智能手机嵌入了各种惯性传感器,如加速度计、陀螺仪和磁力计等。因此,基于智能手机的行人航迹推算(Pedestrian Dead Reckoning, PDR)技术逐渐成为室内定位领域的研究热点。当前,该类技术的研究与应用主要面临着一个挑战：为了精确地实现PDR,行人身上手机的放置姿态和位置需要被施加诸多约束。然而,在现实中,手机的姿态和位置会发生任意地变化。因此,针对这一挑战,本文从以下三个方面进行了重点研究。首先,利用六阶巴特沃斯滤波器对传感器原始信号进行低通滤波处理,并通过多传感器数据融合的卡尔曼滤波解算手机姿态,进而构造旋转矩阵实现不稳定的手机坐标系到稳定的大地坐标系的变换。其次,基于大地坐标系的垂直方向加速度,本文提出了峰值检测和零穿越相结合的改进式非约束步数探测算法。所提出的算法能够在行人行走过程中,手机位置发生变化的情况下精确地实现步数探测,正确率达到94.38%。最后,基于大地坐标系的水平方向角速度,本文提出了新颖的基于最小二乘法的非约束航向估计算法。所提出的算法能够在设备航向连续变化的情况下稳定地估算行人航向。在室内环境中,当手机位于摇摆手和前裤口袋时的航向平均绝对误差分别为5.1°和7.1 °。综上所述,本文提出的算法比现有的PDR算法更加健壮和有效,将对PDR在现实生活中的普及起到极大的推动作用。
[Abstract]:In recent years, indoor positioning technology plays a more and more important role in many fields. At the same time, with the development of micromotor system, smart phone embedded a variety of inertial sensors, such as accelerometers, gyroscopes and magnetometers. Therefore, (Pedestrian Dead Reckoning, PDR) technology based on smart phone is becoming a hotspot in indoor positioning field. At present, the research and application of this kind of technology is facing a challenge: in order to accurately realize the position and position of the PDR, pedestrian mobile phone, it is necessary to impose a lot of constraints. However, in reality, mobile phone posture and position will change arbitrarily. Therefore, in view of this challenge, this article has carried on the key research from the following three aspects. First of all, the sixth order Butterworth filter is used to process the original signal of the sensor, and the attitude of the mobile phone is calculated by the Kalman filter of multi-sensor data fusion. Then the rotation matrix is constructed to transform the mobile phone coordinate system into the stable geodetic coordinate system. Secondly, based on the vertical acceleration of geodetic coordinate system, an improved unconstrained step detection algorithm combining peak detection and zero crossing is proposed. The proposed algorithm can accurately detect the number of steps when the mobile phone position changes during walking, and the correct rate is 94.38. Finally, based on the horizontal angular velocity of geodetic coordinate system, a novel unconstrained heading estimation algorithm based on least square method is proposed in this paper. The proposed algorithm can estimate the pedestrian course stably under the condition of continuous variation of the equipment course. In indoor environment, the mean absolute course error of mobile phone is 5.1 掳and 7.1 掳respectively when the mobile phone is in the swinging hand and the front pants pocket. To sum up, the algorithm proposed in this paper is more robust and effective than the existing PDR algorithm, and will play a great role in promoting the popularization of PDR in real life.
【学位授予单位】：内蒙古大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TN929.53
本文编号：2349395
[Abstract]:In recent years, indoor positioning technology plays a more and more important role in many fields. At the same time, with the development of micromotor system, smart phone embedded a variety of inertial sensors, such as accelerometers, gyroscopes and magnetometers. Therefore, (Pedestrian Dead Reckoning, PDR) technology based on smart phone is becoming a hotspot in indoor positioning field. At present, the research and application of this kind of technology is facing a challenge: in order to accurately realize the position and position of the PDR, pedestrian mobile phone, it is necessary to impose a lot of constraints. However, in reality, mobile phone posture and position will change arbitrarily. Therefore, in view of this challenge, this article has carried on the key research from the following three aspects. First of all, the sixth order Butterworth filter is used to process the original signal of the sensor, and the attitude of the mobile phone is calculated by the Kalman filter of multi-sensor data fusion. Then the rotation matrix is constructed to transform the mobile phone coordinate system into the stable geodetic coordinate system. Secondly, based on the vertical acceleration of geodetic coordinate system, an improved unconstrained step detection algorithm combining peak detection and zero crossing is proposed. The proposed algorithm can accurately detect the number of steps when the mobile phone position changes during walking, and the correct rate is 94.38. Finally, based on the horizontal angular velocity of geodetic coordinate system, a novel unconstrained heading estimation algorithm based on least square method is proposed in this paper. The proposed algorithm can estimate the pedestrian course stably under the condition of continuous variation of the equipment course. In indoor environment, the mean absolute course error of mobile phone is 5.1 掳and 7.1 掳respectively when the mobile phone is in the swinging hand and the front pants pocket. To sum up, the algorithm proposed in this paper is more robust and effective than the existing PDR algorithm, and will play a great role in promoting the popularization of PDR in real life.
【学位授予单位】：内蒙古大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TN929.53
【相似文献】
相关硕士学位论文 前1条
1 杨晓坤;面向非约束智能手机的PDR算法研究[D];内蒙古大学;2016年
，本文编号：2349395
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/2349395.html
