地下铲运机自主行驶及卸载的控制研究

发布时间：2018-09-08 08:43

【摘要】：铲运机作为地下矿山开采最重要的运输装备之一,其自主控制技术一直是领域内研究的热门,地下铲运机的自主控制技术主要体现在工作过程的无人化,本文以地下铲运机为研究对象,对其自主行驶及卸载工作过程中所涉及的控制技术进行研究。在分析地下铲运机这种铰接式车辆转向的优缺点后,对铲运机的转向半径和转向曲线进行了研究,重点探索了铲运机在曲线段和直线段之间转向的特性,从自主控制的角度对转向过程中的阻力矩和转向时间进行了推导计算,最后对铲运机的系统动力学进行了分析,建立了转向系统的控制模型,探讨了转向油缸和转向角之间的约束关系,为自主控制提供理论基础。在自主行驶的控制技术研究中,首先介绍了以航迹推测法为主的相对定位和以“伪GPS”为主的绝对定位相融合的组合定位技术,在统一导航坐标系的前提下,提出全局最优路径规划和局部避障最优路径规划相结合的导航策略,对铲运机在巷道内行驶过程中的运动轨迹参数和运动轨迹曲线进行了定义和描述,并对运动轨迹模型进行了推导,得出了运动轨迹参数之间的关系,以偏离位移、偏航角和偏航角的变化趋势等为反馈修正量,建立以多元信息融合为基础,辅以带约束的控制算法模型,通过对行驶速度和铰接角度的实时控制,实现铲运机的自主行驶控制。以MATLAB/Simulink为仿真平台,结合铲运机转向动力学控制模型,建立铲运机自主行驶控制的仿真模型,以全局最优路径和局部避障最优路径规划为前提,对自主行驶过程中跟踪直线、圆以及正弦曲线三种情况分别进行仿真,验证控制算法,结果表明算法能够有效避开路径中存在的障碍物,能够快速、精确、稳定地实现自主行驶的控制。以惯性单元为测量器件,采用四元数法实现铲运机俯仰角、横滚角和航向角的实时求解和在线更新。结合液压阀的固有特性,对铲运机的举升液压系统进行了仿真并对举升压力进行了推算,以大臂姿态角的变化和举升液压压力力的实时改变为信息来源,结合铲斗固有参数推算铲斗内矿石的实时重量,用以控制铲运机的自主卸载过程,最后给出了自主卸载的控制流程。针对自主行驶和卸载控制中对传感器的需求,构建了以里程计、转角传感器、惯性单元为主的信息采集系统,建立了以CAN总线和以太网通讯为主要数据交互方式的通信系统：针对环境自主识别的扫描传感器的测量跳变问题,提出一种基于虚拟扫描测量预处理滤波技术,快速准确的消除较大测量数据误差的影响。最后搭建了基于嵌入式Linux操控系统的车载控制平台,结合最底层的执行控制器介绍了软件设计的流程和编程平台,形成了自主控制的软硬件平台；为了验证铲运机的自主控制,本文以KCY-2型铲运机为试验平台,选择以J1939总线控制为主的电喷柴油发动机,配以APC120控制为主的自动换档变速箱,在地表模拟巷道环境下进行了自主行驶控制的试验,验证了算法的可行性。对实验过程中的各项数据进行了分析和总结,并将此算法用于地下巷道环境的自主行驶试验。最后进行了自主卸载的控制试验,实现了预期目标。
[Abstract]:As one of the most important transportation equipments in underground mining, the self-control technology of LHD is always a hot topic in the field. The self-control technology of underground LHD is mainly embodied in the unmanned working process. This paper takes the underground LHD as the research object, and the control technology involved in its self-driving and unloading process is studied. After analyzing the advantages and disadvantages of the steering of the articulated vehicle of the underground scraper, the steering radius and the steering curve of the scraper are studied. The steering characteristics of the scraper between the curve section and the straight section are emphatically explored. The resistance moment and the steering time in the steering process are deduced and calculated from the angle of autonomous control. Finally, the system dynamics of the LHD is analyzed, the control model of the steering system is established, and the constraint relationship between the steering cylinder and the steering angle is discussed, which provides a theoretical basis for autonomous control. The integrated positioning technology of the master absolute positioning and the local obstacle avoidance optimal path planning is put forward on the premise of the unified navigation coordinate system. The trajectory parameters and the trajectory curves of the scraper are defined and described, and the trajectory is also described. The model is deduced and the relationship between the motion trajectory parameters is obtained. Taking the deviation displacement, the variation trend of the yaw angle and yaw angle as feedback correction variables, a control algorithm model based on multi-information fusion and supplemented by constraints is established. Through real-time control of the driving speed and articulation angle, the autonomous driving control of the LHD is realized. Taking MATLAB/Simulink as the simulation platform and combining with the steering dynamics control model of the scraper, the simulation model of the scraper autonomous driving control is established. On the premise of global optimal path planning and local obstacle avoidance optimal path planning, the following straight line, circle and sinusoidal curve in the process of autonomous driving are simulated respectively to verify the control calculation. The results show that the algorithm can effectively avoid the obstacles in the path, and realize the control of autonomous driving quickly, accurately and steadily. Taking the inertial unit as the measuring device, the quaternion method is used to solve the pitch angle, roll angle and heading angle of the scraper in real time and update them on line. The lifting hydraulic system is simulated and the lifting pressure is calculated. The real-time weight of ore in the bucket is calculated with the change of arm attitude angle and the real-time change of lifting hydraulic pressure as the information source, and the inherent parameters of the bucket are combined to control the independent unloading process of the scraper. To meet the requirements of sensors in autonomous driving and unloading control, an information acquisition system based on odometer, angle sensor and inertia unit is constructed, and a communication system based on CAN bus and Ethernet communication is established. Aiming at the problem of measurement jump of scanning sensors with autonomous environmental identification, a new method based on CAN bus and Ethernet communication is proposed. In the end, a vehicle control platform based on embedded Linux operation and control system is built, and the software design flow and programming platform are introduced with the bottom executive controller, forming an autonomous control software and hardware platform. In this paper, KCY-2 type scraper is used as the test platform, and J1939 bus-controlled diesel engine is selected as the main engine, and APC120-controlled automatic gearbox is used as the main transmission. The experiment of autonomous driving control is carried out in the simulated tunnel environment. The feasibility of the algorithm is verified. According to the analysis and summary, the algorithm is applied to the autonomous driving test of underground roadway environment. Finally, the control experiment of autonomous unloading is carried out and the expected goal is achieved.
【学位授予单位】：北京科技大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TD422.4

【相似文献】