基于MEMS的精确计步算法的设计与实现

计步已经成为各种穿戴设备的基本功能,但较难实现对任意大小步长距离的精确计算,为此,设计了一种新型的计步器系统,系统硬件上采用了MPU6050传感器模块和CC2541蓝牙模块,并将传感器平置于鞋底.软件上设计了上位机,可接收MPU6050采集的足部运动数据,并计算加速度、实现坐标系变换.提出了一种脉冲提取的算法,算法能够分辨行走状态与静止状态,并提取行走状态下的加速度值,对加速度值采用积分原理计算单步步长.经测试,算法能够将任意距离步长的测量误差控制在3.4%以内.

Design and implementation of accurate step-counting algorithm based on MEMS

Step-counting has become a basic function of all kinds of wearable electronic equipment,but it's difficult to precisely calculate a random step distance.In order to overcome this difficulty,a new type of pedometer system is designed,this system adopts MPU6050 sensor module and CC2541 Bluetooth module in hardware,and put the sensor into the bottom of shoe horizontally. In the aspect of software,an upper personal computer is designed to receive data of foot movement acquired by MPU6050,it also can calculate accelerations and realize coordinate system transform.In addition to the above basic operations,an algorithm for pulse extraction is proposed to distinguish walking state and static state,and to extract acceleration values under the walking state,those acceleration values by theory of integral to calculate length of each step. After testing,the algorithm is able to control measurement error of any distance of step within 3.4%.