Kalman滤波算法在自平衡机器人中的应用

针对两轮自平衡机器人系统的单一惯性传感器采集数据不够准确,且极易受到外界噪声信号的干扰的问题,提出了基于片上可编程系统的卡尔曼滤波算法来实现多惯性传感器的数据融合系统.介绍了卡尔曼滤波器的基本原理、特点和应用环境.利用FPGA控制器的硬件可重构特性,搭建了Nios软核处理器的SOPC系统硬件平台,在硬件平台上采用C语言实现多姿态传感器的卡尔曼滤波算法,为两轮自平衡机器人的多传感器姿态数据融合提供了有效工具,获得了自平衡机器人姿态数据的最优估计,解决了陀螺仪和加速度计惯性传感器的数据补偿问题.测试结果表明,采用FPGA硬件平台实现卡尔曼滤波算法效率高,姿态数据融合准确、可靠,能够满足自平衡机器人控制系统的姿态最优估计和倾角数据实时反馈要求,系统工作稳定.

Kalman Filter Algorithm Realization Based on SOPC in Self-balancing Robot

Two wheeled self-balancing robot system for a single inertial sensor data collection is not accurate enough,and vulnerable to external noise interference.This paper introduces Kalman（Kalman） filtering algorithm based on SOPC to realize multi-inertia Sensor data fusion system design.Basic principles of Kalman filter,characteristics and applications are introduced.Using reconfigurable FPGA features,Nios soft core processor SOPC system hardware platform was built.The hardware platform used C language for more Attitude Sensor Kalman filter algorithm,which provided an effective tool for the two wheeled self-balance robot multi-sensor data fusion to obtain the data from the balance of the optimal estimation of robot pose and solve the inertial gyroscope and accelerometer sensor data compensation.Test results show that the FPGA hardware platform Kalman filter,are highly efficient,attitude data fusion accurate,reliable and able to meet the self-balancing robot control system estimates the optimal attitude angle data in real-time feedback and requirements,and the system is stable.