面向精准农业应用的家机导航关键技术研究
文献类型：学位论文
| 作者 | 高雷
|
| 学位类别 | 博士 |
| 答辩日期 | 2014-05-16 |
| 授予单位 | 中国科学院沈阳自动化研究所 |
| 导师 | 胡静涛 |
| 关键词 | 农机导航 转向控制 导航控制 ISO11783协议 运动预测模型 模糊自适应比例控制 |
| 其他题名 | Key Technologies Study of Agricultural Machinery Navigation in Precision Agriculture Application |
| 学位专业 | 机械电子工程 |
| 中文摘要 | 进入21世纪以来，随着农业生产模式从传统的粗放作业模式向先进的精准作业模式转变,精准农业技术在欧美发达国家已经得到了广泛的推广应用，农作物生产的成本降低了20%以上，经济效益显著。农机导航技术是实现农业精准作业的核心技术之一，贯穿于种植、喷洒、收获的全过程，是精准农业技术的一种基本应用。农机导航技术通过位姿传感装置，获取农机的当前位置和姿态信息，将其与目标路径对比，根据结果控制农机的行进方向，使其沿目标路径行驶。由于能够自动控制农机的行驶方向，所以驾驶员可以将注意力集中在作业机具的操作上，显著降低了作业强度、提高了生产效率和农机利用率。农机导航控制研究中需要解决的共性问题包括：（1）设计开发一个通用的农机导航控制平台及相关装置，可以方便地应用在各种农机上；（2）转向控制方法的研究；（3）农机模型及导航控制方法的研究；（4）在不同的农机上从硬件、软件、网络角度，如何快速集成安装导航控制系统。本文对上述四个问题进行了深入研究，设计了通用导航控制平台，提出了相关模型和控制算法，在联合收割机和拖拉机上成功地进行了平台的集成应用，并完成了路面和田间自动导航控制实验。具体的研究内容包括：首先，设计了分布式导航控制平台。根据农机导航控制系统的功能需求，设计了基于CAN总线的导航控制平台总体结构，划分了各装置的功能，给出了整个导航控制平台的工作原理。对系统中的各主要装置：田间计算机、导航控制器、机械式转向执行机构、电液式转向执行机构、连杆式转向角测量机构、直线位移式转向角测量机构和多功能通用智能节点等进行了详细的功能及软硬件设计。针对农机自动导航控制中对通信功能的需求，在农林机械串行通信控制网络协议ISO 11783的基础上，自行设计了导航通信协议。第二，进行了转向执行机构的建模与控制方法研究，提出了基于预测控制和PD控制的串级转向控制方法。以安装有机械式转向执行机构的农机为研究对象，对转向执行机构采用机理建模的方法，对农机转向系统采用辨识建模的方法，建立了自动转向控制系统模型。研究设计了串级转向控制方法，内环角速度控制器采用动态矩阵控制方法，外环角度控制器采用PD控制方法。在CarSim与Simulink联合仿真环境中，将提出的转向控制方法进行了仿真验证，并在洋马VP6插秧机上进行了实车实验，验证了所提转向控制方法具有跟踪速度快、精度高的优点。第三，提出了农机运动状态预测模型。在农机二轮车运动学模型基础上，根据几何关系，推导提出了农机运动状态预测模型，并将该模型应用在PID控制方法中进行了直线路径追踪仿真。与传统PID控制方法比较，引入该模型后，能够在不明显降低追踪速度的前提下，大幅减小超调，且稳态追踪误差更小。第四，提出了基于模糊自适应比例控制的农机导航控制方法。由于传统比例控制方法中采用恒定的比例系数，则当偏差较大时，追踪速度会比较慢，当偏差较小时依然会进行较为频繁的调节。所以本文提出了基于模糊自适应比例控制的导航控制方法，根据横向偏差和航向偏差的值，采用模糊推理在线调节比例系数，在提高路径追踪速度的同时也降低了追踪误差。第五，制定了农机导航控制系统集成规范。针对目前国内研究开发的农机导航控制系统结构多样、接口封闭、难以与农机其它车载电子系统互联等缺点，研究提出了农机导航控制系统的集成规范。从硬件系统集成、软件系统集成和网络集成三个方面，对农机导航控制系统的集成技术进行了研究，制定了相关接口规范。最后，上述研究内容的成果，构成了一套通用的农机导航控制平台和控制方法，并成功地应用在联合收割机和拖拉机上。其中，在雷沃谷神联合收割机上采用机械式转向执行机构进行转向控制，在雷沃欧豹TA800拖拉机上采用电液式转向执行机构进行转向控制，并且都成功地完成了路径追踪实验，追踪精度能够满足农业精准作业对自动导航控制精度的要求。 |
| 索取号 | S232.3/G25/2014 |
| 英文摘要 | In the 21st century, with the agricultural production pattern shifting from the traditional extensive mode to the advanced precise mode, precision agriculture technology has been widely applied in developed countries. The cost of crop production reduced more than 20%, achieved remarkable economic benefits. Agricultural machinery navigation technology is one of the key technologies to achieve precision agriculture, throughout the the entire process of planting, spraying and harvesting. It is the elementary application of precision agricultural technology.Agricultural machinery navigation technology gets the location and posture information from the position and orientation sensors, then contrasts with the target path, and controls the travel direction of agricultural machine according to the results, in order to make it traveling along the target path. Because the travel direction of agricultural machine can be automatically controlled, the driver can concentrate on the farm implements operating; this decreased the operating strength significantly, and improved agricultural productivity and agricultural machinery utilization.The common problems that need to be addressed in agricultural machinery navigation control study are: (1) design and develop a universal machinery navigation control platform and related apparatus; (2) steering control method Study; (3) agricultural machinery model and navigation control method study; (4) how to quickly integrate and install the navigation control system from the point of view of hardware, software and network on the different agricultural machines. This thesis studies the above four problems, designs the universal agricultural machinery navigation control platform, proposes relating model and control algorithm, and applies integration of the navigation control platform on a combine and a tractor successfully. Both of them accomplished the path tracking experiment successfully. Concrete research content including: First of all, distributed navigation control platform is designed. According to the functional requirements of the agricultural machinery navigation control system, design the overall structure of the navigation control platform based on CAN bus, specified the function of each unit, and described the working principle of the whole navigation control platform. The detailed design of function, hardware and software were accomplished on the main devices in the system: field computer, navigation controller, mechanical steering actuator, electro-hydraulic steering actuator, connecting rod type steering angle measuring mechanism, linear displacement type steering angle measuring mechanism and multi-function universal intelligent node. In order to satisfy the needs of communication function in the operation of agricultural machinery automatic navigation control, on the basis of ISO 11783, navigation communication protocol is designed. Second, the modeling and control method of steering actuator are studied, and the cascade control method based on predictive control and PD control method was proposed. Take the agricultural machinery equipped with mechanical steering actuator as the research object, adopt the mechanism modeling method on steering actuator, and adopt the identification modeling method on agricultural machinery steering system, and an automatic steering control system model is established. The cascade control method is designed. The inner angular velocity controller adopts Dynamic Matrix Control method, and the outer angle controller adopts proportional plus derivative control method. Simulation is made for the proposed steering control method in CarSim associated with Simulink simulation environment, and the actual vehicle experiment was carried out on the YANMAR VP6 transplanter. These simulation and experiment verified that the proposed steering control method has the advantage of fast tracking speed and high precision. Third, the agricultural machinery motion prediction model is proposed. Based on the bicycle model of agricultural machinery, according to the geometric relationship, the agricultural machinery motion prediction model is deduced. The model was applied to the PID control method in a straight path tracking simulation. Compared with traditional PID control method, after introducing this model, the overshoot is sharply reduced; the steady-state tracking error is smaller without remarkably reducing the tracking speed. Fourth, the agricultural machinery navigation control method based on fuzzy adaptive proportional control is proposed. Because of the traditional proportional control method using constant proportion coefficient, when the deviation is big, tracking speed will be slow, when the deviation is small, adjustment still will be relatively frequent. So this paper presents a navigation control method based on fuzzy adaptive proportional control. According to the lateral deviation error and the heading error, this method adopts fuzzy reasoning to adjust proportion coefficient on-line, this method improves the path tracking speed as well as reduces the tracking error.Fifth, an integration specification of agricultural machinery navigation control system is established. Agricultural machinery navigation control systems developed in current researches have a variety of structures, interfaces are closed, and difficult to reconcile with other vehicle electronic systems. Aiming at these shortcomings, an integration specification of agricultural machinery navigation control system is established. From the three aspects, hardware system integration, software system integration and network integration, the integration technologies of agricultural machinery navigation control system are studied, and the relevant interface specification is formulated. Finally, the research contents above constitute a set of universal agricultural machinery navigation control platform and control methods, and successfully applied in a combine and a tractor. Among them, the Lovol Gushen combine adopts mechanical steering actuator for steering control, and the Lovol Oubao TA800 tractor adopts electro-hydraulic steering actuator for steering control. Both of them accomplished the path tracking experiment successfully, and the automatic navigation control precision has met the requirement of precision agriculture. |
| 语种 | 中文 |
| 产权排序 | 1 |
| 页码 | 110页 |
| 分类号 | S232.3 |
| 源URL | [http://ir.sia.ac.cn/handle/173321/14807]  |
| 专题 | 沈阳自动化研究所_信息服务与智能控制技术研究室 |
| 推荐引用方式 GB/T 7714 | 高雷. 面向精准农业应用的家机导航关键技术研究[D]. 中国科学院沈阳自动化研究所. 2014. |
入库方式： OAI收割
来源：沈阳自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。