基于视觉信息丢失的机器鱼运动控制算法

谢景鹏; 李宗刚; 杜亚江; 葛立明

doi:10.11993/j.issn.2096-3920.2020.05.006

基于视觉信息丢失的机器鱼运动控制算法

doi: 10.11993/j.issn.2096-3920.2020.05.006

谢景鹏^1,2,
李宗刚^1,2,
杜亚江^1,2,
葛立明^1,2

1.
兰州交通大学机电工程学院, 甘肃兰州, 730070
2.
兰州交通大学机器人研究所, 甘肃兰州, 730070

基金项目: 国家自然科学基金项目(61663020) 甘肃省高等学校科研项目成果转化项目(2018D-10).

详细信息

作者简介:
谢景鹏(1993-), 男, 硕士, 主要研究方向为多智能体和机器人.

中图分类号: TP242.62
计量
- 文章访问数: 215
- HTML全文浏览量: 27
- PDF下载量: 120
- 被引次数: 0
出版历程
- 收稿日期: 2019-09-17
- 修回日期: 2020-05-25
- 刊出日期: 2020-10-31

Robotic Fish Motion Control Algorithm Based on Visual Information Loss

1.
College of Mechanical Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China
2.
Institute of Robotics, Lanzhou Jiaotong University, Lanzhou 730070, China

摘要

摘要: 针对多机器鱼运动控制中视觉信息丢失的情况, 提出了一种运动补偿控制算法。首先, 针对机器鱼识别状态, 阐述了一种基于事件频率的二次滤波法判定策略。在此基础上建立了半闭环控制的直线追踪法数学模型, 通过对动作矩阵的计算和规划, 从而实现机器鱼近似直线的运动控制。然后基于时间-角度关系, 对圆弧切线法进行了推导, 进而实现在时间域内对机器鱼转弯的运动控制。仿真和实验结果均表明所提算法可以解决在视觉丢失问题下仿生机器鱼的运动控制问题。
- 机器鱼 /
- 视觉信息丢包 /
- 运动补偿控制 /
- 二次滤波 /
- 直线追踪法 /
- 圆弧切线法
Abstract: To solve the problem of visual information loss in multi-robotic fish motion control, a motion compensation control algorithm is proposed. First, for the recognition status of a robotic fish, a judgment strategy using the secondary filtering method based on event frequency is expounded. Subsequently, the mathematical model of the straight-line tracking method with semi-closed loop control is established. Through the calculation and planning of an action matrix, the approximate linear motion control of the robotic fish is realized. Subsequently, based on the time–angle relationship, the circle tangent method is derived to realize the turning motion control of the robotic fish in the time domain. Simulation and experimental results show that the proposed algorithm can solve the motion control problem of bionic robotic fish when visual information loss occurs.
- robotic fish /
- visual information packet loss /
- motion compensation control /
- secondary filtering /
- straight-line tracking method /
- circle tangent method

HTML全文

参考文献(10)

[1]	Ni X, Wang X, Wang S, et al. Structure Design and Image Recognition Research of a Picking Device on the Apple Picking Robot[J]. IFAC Papers on Line, 2018, 51(17): 489-494.
[2]	徐亮. 基于云计算平台的机器鱼目标识别并行深度学习研究[D]. 北京: 中国地质大学, 2018.
[3]	王培屹. 基于多传感器多目标实时跟踪视觉系统在全自主机器人上的应用[J]. 软件导刊, 2011, 10(1): 111-113.
[4]	韩珍, 颜钦, 张世武, 等. 基于序列图像处理法的机器鱼转向机动性能研究[J]. 机器人, 2010, 32(5): 586-591. Han Zhen, Yan Qin, Zhang Shi-wu, et al. Study on Turning Maneuverability of Robotic Fish with Sequential Image Processing[J]. Robot, 2010, 32(5): 586-591.
[5]	罗朝辉, 李丽宏, 谭爽. 基于图像处理的机器人控制系统的研究设计[J]. 现代制造工程, 2018(10): 65-71. Luo Zhao-hui, Li Li-hong, Tan Shuang. Research and Design of Robot Control System Based on Image Processing[J]. Modern Manufacturing Engineering, 2018(10): 65-71.
[6]	Lulio L C, Tronco M L, Porto A J V. Cognitive-merged Statistical Pattern Recognition Method for Image Processing in Mobile Robot Navigation[C]//The Proceedings of the 2012 Brazilian Robotics Symposium and Latin American Robotics Symposium. Fortaleza, Brazil: IEEE, 2012: 279-283.
[7]	杨光, 周鹏举, 张宋彬, 等. 基于卷积神经网络的变电站巡检机器人图像识别[J]. 软件, 2017, 38(12): 190-192. Yang Guang, Zhou Peng-ju, Zhang Song-bin, et al. Image Recognition of Substation Inspection Robot Based on Convolutional Neural Network[J]. Computer Engineering & Software, 2017, 38(12): 190-192.
[8]	Yu J, Liu L, Wang L, et al. Turning Control of a Multilink Biomimetic Robotic Fish[J]. IEEE Transactions on Robotics, 2008, 24(1): 201-206.
[9]	Sun F, Yu J, Chen S, et al. Active Visual Tracking of Free-swimming Robotic Fish Based on Automatic Recognition[C]//The Proceedings of the 11th World Congress on Intelligent Control and Automation. Shenyang, China: IEEE, 2015: 2879-2885.
[10]	唐宇峰. 面向全局视觉机器鱼的多目标跟踪系统的研究与实现[D]. 上海: 华东交通大学, 2013.