董胜, 袁朝辉, 谷超, 杨芳. 基于多学科技术融合的智能农机控制平台研究综述[J]. 农业工程学报, 2017, 33(8): 1-11. DOI: 10.11975/j.issn.1002-6819.2017.08.001
    引用本文: 董胜, 袁朝辉, 谷超, 杨芳. 基于多学科技术融合的智能农机控制平台研究综述[J]. 农业工程学报, 2017, 33(8): 1-11. DOI: 10.11975/j.issn.1002-6819.2017.08.001
    Dong Sheng, Yuan Zhaohui, Gu Chao, Yang Fang. Research on intelligent agricultural machinery control platform based on multi-discipline technology integration[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2017, 33(8): 1-11. DOI: 10.11975/j.issn.1002-6819.2017.08.001
    Citation: Dong Sheng, Yuan Zhaohui, Gu Chao, Yang Fang. Research on intelligent agricultural machinery control platform based on multi-discipline technology integration[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2017, 33(8): 1-11. DOI: 10.11975/j.issn.1002-6819.2017.08.001

    基于多学科技术融合的智能农机控制平台研究综述

    Research on intelligent agricultural machinery control platform based on multi-discipline technology integration

    • 摘要: 农业机械的自动化和智能化包含内容广泛,有农机定位与导航,动态路径规划,机器视觉和远程监控等,牵涉到大量的工程技术学科,包括导航、图像、模型与策略、执行器以及数据链等。农机定位与导航一般采用基于农机运动学模型结合GPS(global positioning system)/IMU(inertial measurement unit)组合导航信息,在导航路径规划算法指引下实现农机轨迹跟踪的方法。建立的农机运动学模型精度,GPS数据的连续性以及惯导器件误差系数漂移等因素都会影响该方法的有效性。路径跟踪通常采用各种现代控制理论与方法,而面对复杂的田间作业环境变化,农机的自主避障以及动态路径规划能力也会影响轨迹跟踪精度。机器视觉的稳定性和目标特征信息分离度影响着农机环境感知能力,目前目标识别主要采用hough变换,hough变换的全局检测特性决定了该算法运算量较大,需要探究改进特征提取算法。远程监控农机作业是智能农机发展的一个方向,构建无线导航,控制和视频数据传输网络有助于提高农机的智能化水平,可以采用分布式哈希表(distributed hash table)来研究网络覆盖和互联技术。该文融合多个学科,从高精度定位与导航技术、复杂环境及工况下农机运动精确自主控制技术、稳定清晰的机器视觉感知技术和基于4G网络和新一代物联网的高覆盖数据传输技术几个方面,论述了智能农机在光机电液多个学科领域内的研究现状,并指出采用北斗地基增强网络和网络RTK(real-time kinematic)技术、惯导定位误差精确建模与补偿、环境感知与自主避障、立体结构自组网技术以及多机协作是现代农业机械的发展方向。以期为现代化智能农业机械的设计提供参考。

       

      Abstract: Abstract: The meaning of agricultural machinery automation and intellectualization includes a wide range of content, involving a large number of engineering disciplines, such as navigation, images, models and strategies, actuators and data chain. The intellectualization of agricultural machinery is the symbol of a country's engineering and technical strength. How to co-ordinate the multidisciplinary technologies and reasonably integrate them into a system is the key to the success of intelligent agricultural machinery. According to the requirements of the development of land scale management in China, intelligent agricultural machinery should use advanced technology in the aeronautics, astronautics and ground to achieve high-speed computing and transmission of data information in a dynamic environment, and to command the actuator to complete the tasks efficiently and effectively. In recent years, the research on intelligent agricultural machinery has been mainly concentrated on vision measuring, image processing, trajectory tracking and vehicle navigation, pattern recognition and their application, and so on, which have focused on image information acquisition method, image processing and recognizing algorithm, intelligent navigation algorithm and system integration application, and so on. The technology of automatic steering control, obstacle detection and active obstacle avoidance, and multi machine cooperative navigation, and the technology of agricultural machinery will be the focus of the next generation of intelligent agricultural machinery. This paper studied the Beidou enhanced network and network RTK (real-time kinematic) method to improve the accuracy of positioning and navigation technology, accurately modeled and compensated aiming at the inertial navigation error of agricultural machinery, and prolonged the working time of positioning. The dynamic model of agricultural machinery was established, and the parameters of the model were identified on the basis of the actual situation. Based on the constraint conditions, the knowledge learning and decision control technology was introduced into the agricultural machinery. And then combined with machine vision, the perception of the working environment and operating objects was enhanced, and the image recognition algorithm was used to assist navigation and control. Finally, according to the requirements of remote monitoring, and navigation and positioning for data transmission, the realization of data transmission technology of three-dimensional topology was studied so as to ensure the real-time data fusion of various technical units. The invention disclosed a set of intelligent device with electromechanical integration. It was equipped with a multi-source fusion intelligent controller based on the dynamic model of agricultural machinery, and an inertial navigation system, which could effectively isolate the disturbance of agricultural machinery to achieve stable tracking and the combination of WIFI and the new generation of Internet of Things to complete the multi-link data communication. According to the dynamic analysis of agricultural machinery, the data model was obtained. The model structure was determined, but the parameters were uncertain. The model parameters could be obtained by on-line identification on the basis of the minimum value of the difference between the actual state and the model state. Then the control strategy was designed according to the model. Based on the navigation control error prediction, the control law parameters were optimized through minimizing the objective function of the future control deviation, and the trajectory tracking of agricultural machinery would be always controlled with the best control parameters. Agricultural navigation used the tight integrated navigation technology of Beidou plus IMU (inertial measurement unit), and machine vision was taken as an aid. Based on the enhanced technology of mobile network base station of Beidou Foundation and high precision standard inertial navigation system for precision agriculture applications, the continuity and reliability of navigation and positioning information were ensured, which laid the foundation for the use of intelligent agricultural machinery in remote sensing areas. Therefore, the agricultural machinery in the next period will be more intelligent and easy in operation. New farmers in the Farm Hall will command all kinds of agricultural machines to carry out operations in a variety of environmental conditions, and monitor agricultural machinery in a few kilometers or tens of kilometers away in real time, which greatly reduce the burden on farmers, and achieve high efficiency, standardization and hommization, and also provide some technical references for the development of other industries.

       

    /

    返回文章
    返回