室外移动机器人的惯导/ 网络RTK组合自定位研究  

Inertial/Network RTK Combined Positioning Researched on Outdoor Mobile Robot

作　　者：鲁烨超 梁冬泰[1] 梁丹 刘鹏翔 LU Yechao;LIANG Dongtai;LIANG Dan;LIU Pengxiang(Faculty of Mechanical Engineering and Mechanics,Ningbo University,Ningbo Zhejiang 315211,China)

机构地区：[1]宁波大学机械工程与力学学院,浙江宁波315211

出　　处：《机械设计与研究》2020年第5期12-17,23,共7页Machine Design And Research

基　　金：国家自然科学基金青年基金(51805280);宁波市自然科学基金(2019A610158)资助项目;宁波市“科技创新2025”重大专项(2019B10027)。

摘　　要：针对移动机器人在室外复杂环境下单一导航模式无法提供精确稳定的定位信息问题,提出了一种基于惯导/网络RTK(Real-time kinematic)的组合自定位算法。首先,以一种模糊自适应互补滤波姿态解算算法,根据加速度计测量值的误差能反映载体运动状态这一特性设计了模糊控制器,实时调整滤波PI控制器的比例参数,进行姿态解算。然后,将惯导与网络RTK提供的载体三维位置和速度信息通过坐标轴转换到导航系下,利用卡尔曼滤波器融合两者导航参数得到机器人的真实位置估计。最后,利用STM32嵌入式开发平台,建立移动机器人硬件平台进行实验,实验结果表明本文的组合自定位算法具有较高的精度且在网络RTK失效(短时间)时能维持正常导航定位。Aiming at the problem that a single navigation mode of mobile robot can not provide accurate and stable positioning information in outdoor complex environment,a combined self positioning algorithm based on inertial navigation/Network RTK(real-time kinematic)is proposed.First,a fuzzy adaptive complementary filter algorithm is proposed in this paper.According to the fact that the measurement error of accelerometer can reflect the motion state of the carrier,a fuzzy controller is designed to adjust the proportional parameters of the filter PI controller in real time to solve the attitude.Then,the three-dimensional position and velocity information provided by ins and RTK are transformed into the navigation system through the coordinate axis,and the real position estimation of the robot is obtained by integrating the navigation parameters of the two systems with Kalman filter.Finally,using STM32 embedded development platform,the hardware platform of mobile robot is established for experiment.The experimental results show that the combined self positioning algorithm has high accuracy and can maintain the normal navigation and positioning when the network RTK fails(short time).

关 键 词：移动机器人 网络RTK 组合定位 模糊自适应控制 卡尔曼滤波 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]