基于多学科技术融合的智能农机控制平台研究综述

发布时间：2018-07-10 03:17

  本文选题：农业机械 + 导航　； 参考：《农业工程学报》2017年08期

【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]:The automation and intelligence of agricultural machinery include a wide range of contents, including agricultural machinery positioning and navigation, dynamic path planning, machine vision and remote monitoring, involving a large number of engineering and technical disciplines, including navigation, images, models and strategies. Actuators and data links. Based on the kinematics model of agricultural machinery and the integrated navigation information of (global positioning system) / IMU (inertial measurement unit), agricultural machinery location and navigation are usually implemented under the guidance of navigation path planning algorithm. The precision of the established kinematics model and the continuity of GPS data, as well as the error coefficient drift of inertial navigation devices, will affect the effectiveness of the method. Path tracking usually adopts various modern control theories and methods, but in the face of complex field operation environment changes, the ability of autonomous obstacle avoidance and dynamic path planning of agricultural machinery will also affect the track tracking accuracy. The stability of machine vision and the separation degree of target feature information affect the environmental perception ability of agricultural machinery. At present, the global detection characteristics of hough transform and Hough transform are the main methods for target recognition. The improved feature extraction algorithm needs to be explored. Remote monitoring of agricultural machinery operation is a direction of intelligent agricultural machinery development. Building wireless navigation, control and video data transmission network will help to improve the intelligent level of agricultural machinery. Distributed hash table (distributed hash table) can be used to study network coverage and interconnection technology. This paper combines many disciplines, including high precision positioning and navigation technology, precise autonomous control technology of agricultural machinery movement under complex environment and working conditions. The stable and clear machine vision perception technology and the high coverage data transmission technology based on 4G network and the new generation of Internet of things are discussed in this paper. It is pointed out that the technology of beidou foundation enhancement network and network RTK (real-time kinematic), accurate modeling and compensation of inertial navigation positioning error, environment perception and autonomous obstacle avoidance, stereo structure ad hoc network technology and multi-machine cooperation are the development direction of modern agricultural machinery. In order to provide reference for the design of modern intelligent agricultural machinery.
【作者单位】： 西北工业大学自动化学院;飞行器控制一体化技术重点实验室中航工业自控所;河南科技大学机电学院;
【基金】：航空科学基金(MIMU/单星伪距伪距率组合导航技术(20165853041))
【分类号】：S126;S220
，

本文编号：2111822