移动机器人的室内地图创建及其实现

发布时间：2018-06-17 18:23

本文选题：移动机器人 + SLAM　；参考：《西南科技大学》2017年硕士论文

【摘要】：在机器人研究领域中,同时定位与地图创建(SLAM,Simultaneous Localization and Mapping)技术是机器人实现自主导航的关键,其中,研究地图创建方法具有重大意义。本文针对室内SLAM技术的关键问题进行了深入研究与分析,建立了相应的问题求解模型;同时,结合实际应用需要,在机器人建图过程中引入了多传感器数据融合和激光扫描匹配的方法提高了机器人定位的准确性,并且建立了栅格反射率计算模型更新栅格的占据概率,提高了地图创建的精度和效率;最后,搭建了实际的机器人实验平台并验证了改进算法的可行性和有效性。针对实际应用中码盘里程数据精度低引起的机器人姿态角不准确的问题,本文采用扩展卡尔曼滤波的方法融合码盘数据和陀螺仪数据,以提高机器人位姿测量的精度,保证算法执行的有效性。针对传统的里程计定位方法中里程数据的累积误差引起的机器人位姿不准确的问题,本文采用扫描匹配的方法求取激光雷达的连续测量数据之间的相对位姿变换,以此对里程计定位结果进行有效校正,提高了机器人长距离定位的精度。针对传统的占据栅格地图创建方法中计算复杂、地图边缘效果差的问题,本文采用栅格反射率的计算方法更新栅格的占据概率,提高了栅格地图创建的精度和计算效率。为验证本文所述算法的实用性和可行性,搭建了一个实际的机器人实验平台,实现了相关的SLAM功能,并通过实验数据分析验证了其有效性。
[Abstract]:In the field of robot research, the technology of simultaneous localization and map creation is the key of robot autonomous navigation, and it is of great significance to study the method of map creation. In this paper, the key problems of indoor slam technology are deeply studied and analyzed, and the corresponding problem solving model is established. The method of multi-sensor data fusion and laser scanning matching is introduced to improve the accuracy of robot location, and the raster reflectivity calculation model is established to update the occupied probability of grid. The accuracy and efficiency of map creation are improved. Finally, an actual robot experimental platform is built and the feasibility and effectiveness of the improved algorithm are verified. In order to improve the precision of robot pose measurement, the extended Kalman filter method is used to fuse the data of code disk and gyroscope, which is caused by the low precision of robot attitude angle caused by the low precision of code disk mileage data in practical application, in order to improve the accuracy of robot pose measurement. Ensure the effectiveness of the algorithm execution. Aiming at the inaccuracy of robot pose caused by the accumulated error of mileage data in the traditional odometer positioning method, this paper uses the scanning matching method to calculate the relative position and attitude transformation between the continuous measurement data of lidar. With this method, the positioning results of the odometer are corrected effectively, and the precision of the long distance positioning of the robot is improved. In order to solve the problem of complex calculation and poor edge effect in the traditional method of constructing occupied raster map, this paper uses the method of raster reflectivity to update the occupied probability of raster, which improves the accuracy and efficiency of raster map creation. In order to verify the practicability and feasibility of the algorithm described in this paper, a practical robot experimental platform is built, and the related slam function is realized, and the validity of the platform is verified by the analysis of experimental data.
【学位授予单位】：西南科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242

【相似文献】