大型运载车平台自主定位关键技术研究

发布时间：2018-07-06 18:22

本文选题：运载车 + 激光雷达　；参考：《上海交通大学》2012年硕士论文

【摘要】：大型运载车平台是一种广泛用于重型设备运输的特殊装备。随着相关工程机械技术的不断进步和工程建设需求的提高,大型运载车等装备的研究开始朝着智能化的方向发展。大型运载车的自主定位技术就是其中的一个重要内容。而普遍的基于GPS,惯性组合等方法的车辆定位由于其自身的限制,并不适用于稳定性和精确性等要求较高的大型运载车定位环境,因此结合大型运载车平台特点的自主定位技术研究对大型运载设备的智能化发展有着重要意义。本文在对运载车平台结构及其定位特性分析的基础上,提出了基于激光雷达特征识别和信息融合技术的自主定位方法。本文的研究主要包含三部分内容: 1.基于对LMS111激光雷达测量原理和数据特点的研究分析,应用聚类-拟合算法实现了信标特征识别。该算法的基本思想是根据激光雷达二维扫描的距离图像对原始测量数据进行聚类分组,然后分别进行线段的正交回归拟合与圆弧检测,按照一定的匹配准则进行组合信标匹配,最后进行特征识别实验对该算法的性能进行了验证。 2.建立了运载车定位系统的位姿估计模型。根据运载车的定位系统工作原理分析,建立了系统的状态及观测模型。基于激光雷达预处理数据和运载车转向角反馈信息,应用扩展卡尔曼滤波器实现了多信息融合的位姿估计,通过室内模拟实验对该方法的稳定性及其对随机噪声的抑制进行了验证。 3.完成了定位系统实现与整车定位实验。基于研究验证的自主定位方法,本文设计了运载车自主定位系统的结构方案,并结合实际的大型运载车平台完成了系统定位功能、人机交互、系统监控等模块的软硬件实现。最终在该系统上进行了全过程的自主定位实验,大量的实验结果表明了本文所述的自主定位技术方案的可行性和可靠性,为今后大型运载车自主定位技术的深入研究提供了一定的借鉴。
[Abstract]:Large vehicle platform is a special equipment widely used in heavy equipment transportation. With the continuous progress of related construction machinery technology and the improvement of engineering construction demand, the research of large vehicles and other equipment began to develop intelligently. The autonomous positioning technology of large vehicle is one of the important contents. However, the general vehicle positioning based on GPS, inertial combination and other methods is not suitable for large vehicle positioning environment with high stability and accuracy due to its own limitations. Therefore, the research of autonomous positioning technology based on the characteristics of large vehicle platform is of great significance to the intelligent development of large-scale transportation equipment. Based on the analysis of the platform structure and location characteristics of the carrier vehicle, this paper presents an autonomous localization method based on lidar feature recognition and information fusion technology. This paper mainly includes three parts: 1. Based on the research and analysis of LMS111 lidar measurement principle and data characteristics, the feature recognition of beacon is realized by clustering and fitting algorithm. The basic idea of this algorithm is to cluster and group the original measured data according to the range image of the two dimensional scanning of lidar, and then carry out the orthogonal regression fitting and arc detection of the line segments respectively, and carry out the combined beacon matching according to certain matching criteria. Finally, the performance of the algorithm is verified by feature recognition experiment. 2. The position and attitude estimation model of vehicle positioning system is established. Based on the analysis of the working principle of the vehicle positioning system, the state and observation model of the system is established. Based on the pre-processing data of lidar and the steering angle feedback information of vehicle, an extended Kalman filter is used to estimate the position and attitude of multi-information fusion. The stability of the method and its suppression of random noise are verified by indoor simulation experiments. The realization of the positioning system and the vehicle positioning experiment are completed. Based on the research and verification of the autonomous positioning method, this paper designs the structural scheme of the autonomous positioning system of the vehicle, and realizes the hardware and software of the system positioning function, man-machine interaction, system monitoring and other modules combined with the actual large-scale vehicle platform. Finally, the whole process of autonomous localization experiment is carried out on the system. A large number of experimental results show the feasibility and reliability of the proposed autonomous positioning technology. It provides a certain reference for the further study of autonomous positioning technology of large vehicles in the future.
【学位授予单位】：上海交通大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH22

【参考文献】