SIA OpenIR  > 水下机器人研究室
面向海洋锋面跟踪的多AUV路径规划方法研究
Alternative TitleResearch on multi-AUV path planning method for ocean front tracking
曲向宇1,2
Department水下机器人研究室
Thesis Advisor李一平
Keyword多AUV 海洋锋面跟踪 在线路径规划 队形控制 温度动态估计
Pages79页
Degree Discipline控制理论与控制工程
Degree Name硕士
2020-05-26
Degree Grantor中国科学院沈阳自动化研究所
Place of Conferral沈阳
Abstract随着人类对海洋资源依赖的程度逐渐加大,人类对各类海洋特征的观测研究也日益增加,其中海洋锋面与人类渔业、海洋军事和科考活动关系密切,将海洋锋面作为海洋观测对象具有重大的研究价值和意义。由于海洋锋面具有较高的水平温度梯度,呈狭长带状空间分布结构,易受洋流、季风、太阳辐射等因素影响而发生变化等特点,可以利用自主水下机器人(Autonomous Underwater Vehicles, AUV)作为海洋锋面跟踪的移动平台,充分发挥AUV高自主性、高航速等特点,实现对海洋锋面的连续跟踪和快速观测。本文的主要研究内容和成果如下:(1)针对海洋锋面跟踪的研究背景,搭建了基于卫星遥感数据的海洋锋面动态温度模拟环境。该方法通过获取到的卫星海温数据,可以对指定时间范围内的观测区域进行温度环境的模拟,为使用AUV进行海洋锋面跟踪算法的验证提供了仿真环境。(2)针对海洋锋面跟踪的需求,提出了一种基于温度动态估计的AUV海洋锋面跟踪算法,该方法通过AUV对已获取到的温度数据进行筛选,实时预测观测半径内的温度数据,并根据观测状态选择不同的路径规划策略,若为搜寻和跟踪状态,则沿梯度方向跟踪来搜索锋区梯度极值点,实现海洋锋面的快速穿越,若为脱离状态,则沿等温线航行来保持对海洋锋面边界的跟踪能力,通过调节边界跟踪的长度,可以实现对海洋锋面的不同精细程度观测。在搭建的锋区温度环境中对算法进行仿真,结果显示,本文算法较于常规算法,在跟踪准确程度和环境适应能力上有更好的表现。(3)针对多AUV进行海洋锋面跟踪时的队形控制问题,研究了基于跟随领航者的队形控制方法,设计了基于运动学模型的队形控制器,通过调整编队参数实现队形的变换,并对AUV在锋面跟踪路径下的队形变换过程进行了仿真,分析了领航AUV航向角的变化对队形误差的影响程度,为多AUV在进行海洋锋面跟踪任务时的队形设计提供了指导方案。 (4)在使用多AUV进行海洋锋面跟踪的背景下,研究了多AUV的在线路径规划方法。结合多AUV观测到的数据,提出了基于最近邻的多样本选择方法,提高海洋锋面跟踪的准确程度,增加队形稳定性;通过对求取设定观测范围内梯度极值,确定观测点之间的最优观测距离,提高算法跟踪速度。通过对上述方法进行仿真实验,根据多组对比跟踪算法的实验结果,验证了本文提出的方法具有更快的海洋锋面跟踪速度和较高的队形稳定性。
Other AbstractAs the increasing dependence of human beings on marine resources, the observation and research of all kinds of ocean features are also growing. Among them, the ocean front is closely related to human fishery, marine military and scientific research activities. Therefore, it is of great research value and significance to observe the front. In view of the high horizontal temperature gradient of the front, which has a long and narrow spatial distribution structure, and is prone to dynamic changes due to ocean current, monsoon, solar radiation and other factors, Autonomous Underwater Vehicles (AUV) can be used as the mobile platform for the front observation, which can give full play to the characteristics of AUV, such as high autonomy, high speed and so on, to realize the continuous tracking and rapid observation of ocean fronts. The research and innovation are mainly reported in the following aspects: (1) According to the research background of ocean front observation, a dynamic simulation temperature field of frontal based on satellite remote sensing data is built. This method can simulate the temperature field environment of the observation area within the specified time range through the satellite SST data obtained, which provides a simulation environment for the verification of the related feature tracking algorithm using AUV. (2) In order to meet the demand of ocean front observation, an AUV front tracking algorithm based on dynamic temperature estimation is proposed. In this method, the acquired temperature data is filtered by AUV, and the temperature data within the observation radius is predicted in real time, and different path planning strategies are selected according to the observation state. If it is a searching and tracking state, the gradient extremum is searched along the gradient direction to realize the fast crossing of the ocean front. If it is out of the state, it can maintain the tracking ability of the ocean front boundary by sailing along the isotherm. By adjusting the length of the boundary tracking, it can realize the observation of the ocean front at different fine degrees. The simulation results show that the algorithm has better performance in tracking accuracy and environmental adaptability than the conventional algorithm. (3) Aiming at the formation control problem of multiple AUVs in feature tracking, the formation control method based on following the pilot is studied, and the formation controller based on kinematic model is designed, which can realize the formation transformation by adjusting the formation parameters. The formation transformation process of AUV in front tracking path is simulated, and the influence degree of the change of heading angle of AUV on formation error is analyzed, which provides a guidance scheme for the formation design of multi AUV in feature tracking. (4) Under the background of front tracking with multiple AUVs, the online path planning method of multi-AUV is studied. Combined with the data observed by multiple AUVs, a multi sample selection method based on the nearest neighbor is proposed to improve the accuracy of front tracking and increase the stability of formation; the optimal observation distance between observation points is determined and the observation speed of the algorithm is improved by calculating the gradient extreme value within the observation range. According to the experimental results of four groups of multi-AUV frontal tracking algorithm, the method in this paper has faster tracking speed and can maintain high formation stability.
Language中文
Contribution Rank1
Document Type学位论文
Identifierhttp://ir.sia.cn/handle/173321/27129
Collection水下机器人研究室
Affiliation1.中国科学院沈阳自动化研究所;
2.中国科学院大学
Recommended Citation
GB/T 7714
曲向宇. 面向海洋锋面跟踪的多AUV路径规划方法研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2020.
Files in This Item:
File Name/Size DocType Version Access License
面向海洋锋面跟踪的多AUV路径规划方法研(5740KB)学位论文 开放获取CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[曲向宇]'s Articles
Baidu academic
Similar articles in Baidu academic
[曲向宇]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[曲向宇]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.