基于场景识别的惯性基类脑导航方法

doi:10.3969/j.issn.1674-5558.2020.h4.014

导航与控制 ›› 2020, Vol. 19 ›› Issue (4/5): 119-125.doi: 10.3969/j.issn.1674-5558.2020.h4.014

基于场景识别的惯性基类脑导航方法

赵菁^1,2, 赵东花^1,2, 王晨光^1,3, 张雨⁴, 申冲^1,2, 唐军^1,2, 刘俊^1,2

1.中北大学电子测试技术国防科技重点实验室,太原 030051;
2.中北大学仪器与电子学院,太原 030051;
3.中北大学信息与通信工程学院,太原 030051;
4.东南大学仪器科学与工程学院,南京 210096

收稿日期:2020-03-14 出版日期:2020-10-05 发布日期:2020-12-22
通讯作者: 申冲,男,汉族,1986年生,博士,中北大学仪器与电子学院副教授,先后入选中北大学青年学术带头人、山西省优秀青年学术带头人。主持国家自然科学基金、中央军委装备发展部领域基金以及山西省科技基金等国家/省部级项目5项,承担中央军委装备发展部十三五预研1项。目前主要从事仿生/智能导航定位方向研究,并在该领域取得一系列创新性研究成果(为本文通信作者,Email:shenchong@nuc.edu.cn)。
作者简介:赵菁,女,1996年生,仪器科学与技术专业,中北大学仪器与电子学院硕士研究生,研究方向为类脑导航、视觉导航。刘俊,男,汉族,1968年生,教授、博士生导师,现任中北大学副校长、中北大学仪器与电子学院院长兼测试技术及仪器实验教学中心主任,为国家杰出青年基金获得者,科技部中青年创新领军人才,全国五一劳动奖状获得者,享受国务院特殊津贴专家。目前主要从事微纳光量子器件与智能测量和仿生传感与惯性导航研究,主持完成国家“973”“863”、国家自然科学基金重大仪器专项/重点、军委科技委JCJQ计划等项目50余项。获国家技术发明二等奖2项、省部级奖12项,发表SCI论文100余篇,授权国家发明专利50余项。
基金资助:
国家自然科学基金(编号:61973281);山西省研究生教育创新项目(编号:2019SY441)

Inertial-based Brain-like Navigation Strategy Based on Scene Recognition

ZHAO Jing^1,2, ZHAO Dong-hua^1,2, WANG Chen-guang^1,3, ZHANG Yu⁴, SHEN Chong^1,2, TANG Jun^1,2, LIU Jun^1,2

1. Science and Technology on Electronic Test and Measurement Laboratory, North University of China, Taiyuan 030051;
2. School of Instrument and Electronic, North University of China, Taiyuan 030051;
3. School of Information and Communication Engineering, North University of China, Taiyuan 030051;
4. School of Instrument Science and Engineering, Southeast University, Nanjing 210096

Received:2020-03-14 Online:2020-10-05 Published:2020-12-22

摘要/Abstract

摘要： 针对无人机导航中惯性器件产生的漂移误差不断随时间累积进而影响导航精度的问题,以仿生类脑导航为研究背景,提出了一种新的导航方法。与传统导航策略不同的是,该方法从周期性校正累积位置误差的角度出发,采用设置位置细胞节点的思路,利用训练好的卷积神经网络模型在细胞节点处进行图像匹配,从而在位置细胞节点处实现惯导位置漂移误差校正。同时,建立节点之间的漂移误差模型,调整误差方程系数,以达到修正漂移误差的目的。最后,基于无人机飞行实验结果验证了该方法对自主导航的有效性和鲁棒性,该方法能够有效提高无人机导航的精度。

关键词: 类脑导航, 图像匹配, 卷积神经网络, 节点误差补偿

Abstract: The drift error generated by the inertial devices is accumulating with time, affecting the accuracy in UAV navigation. To solve that problem, a new navigation method based on the bionic brain-like navigation is proposed. Different from the traditional navigation strategy, this method is inspired by setting position cell nodes in bionic brain-like strategy that trains convolutional neural network model to perform image matching at cell nodes, so as to achieve the inertial navigation position drift error correction at the position cell node. In addition, the drift error model between nodes is established and the coefficient of error equation is adjusted to correct the drift error. Finally, based on the flight test results of UAV, the effectiveness and robustness of the proposed method for autonomous navigation are verified. The accuracy of UAV navigation is effectively improved by the proposed method.

Key words: brain-like navigation, image matching, convolutional neural network, node error compensation

中图分类号:

TP183/V249.32+2

赵菁, 赵东花, 王晨光, 张雨, 申冲, 唐军, 刘俊. 基于场景识别的惯性基类脑导航方法[J]. 导航与控制, 2020, 19(4/5): 119-125.

ZHAO Jing, ZHAO Dong-hua, WANG Chen-guang, ZHANG Yu, SHEN Chong, TANG Jun, LIU Jun. Inertial-based Brain-like Navigation Strategy Based on Scene Recognition[J]. Navigation and Control, 2020, 19(4/5): 119-125.

参考文献

[1] Huang L, Song J M, Chen P H, et al. Visual navigation for UAV using optical flow estimation[C]. Proceedings of the 33^rd Chinese Control Conference, 2014: 816-821.
[2] Rhudy M B, Chao H Y, Gu Y. Wide-field optical flow aided inertial navigation for unmanned aerial vehicles[C]. IEEE/RSJ International Conference on Intelligent Robots and Systems, 2014: 674-679.
[3] Tang Y Z,Hu Y C, Cui J Q, et al. Vision-aided multi-UAV autonomous flocking in GPS-denied environment[J]. IEEE Transactions on Industrial Electronics, 2019, 66(1): 616-626.
[4] Chen Y T, Liu L H, Gong Z Q, et al. Learning CNN topair UAV video image patches[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2017, 10(12): 5752-5768.
[5] Jin Z L, Wang X Z, Moran B, et al. Multi-region scene matching based localisation for autonomous vision navigation of UAVs[J]. Journal of Navigation, 2016, 69(6): 1215-1233.
[6] 陈孟元. 鼠类脑细胞导航机理的移动机器人仿生SLAM综述[J]. 智能系统学报, 2018, 13(1): 107-117.
CHEN Meng-yuan. Overview of mobile robot bionic slam based on navigation mechanism of mouse brain cells[J]. CAAI Transactions on Intelligent Systems, 2018, 13(1): 107-117.
[7] 蒋强卫, 甘兴利, 李雅宁. 基于CNN双目特征点匹配目标识别与定位研究[J]. 无线电工程, 2018, 48(8): 643-649.
JIANG Qiang-wei, GAN Xing-li, LI Ya-ning. Research of target recognition and positioning based on CNN binocular feature point matching[J]. Radio Engineering, 2018, 2018, 48(8): 643-649.
[8] Zhang Y, Lin Y, Sun P, et al. IFCNN:a general image fusion framework based on convolutional neural network[J]. Information Fusion, 2019, 54: 99-118.
[9] 李倜. 基于鼠脑海马认知机理的机器人面向目标的导航模型研究[D]. 北京工业大学, 2016.
LI Ti. Research on the target-oriented robot navigation model based on the cognitive mechanism of rat hippocam-pus[D]. Beijing University of Technology, 2016.
[10] Chen Q Y, Mo H W. A brain-inspired goal-oriented robot navigation system[J]. Applied Science, 2019,9(22): 4869.
[11] Banino A, Barry C, Uria B, et al. Vector-based naviga-tion using grid-like representations in artificial agents[J]. Nature, 2018, 557(7705): 429-433.
[12] Hausler S, Chen Z T, Hasselmo M E, et al. Bio-inspired multi-scale fusion[J]. Biological Cybernetics, 2020, 114: 209-229.

基于场景识别的惯性基类脑导航方法

Inertial-based Brain-like Navigation Strategy Based on Scene Recognition

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

编辑推荐

Metrics

[1]	蔡立明, 李威, 高永发, 张文亮, 张玉强. 基于多特征图像增强深度卷积神经网络的航天用电子元器件分类算法[J]. 导航与控制, 2020, 19(2): 112-119.
[2]	陈艳, 曾庆化, 王云舒, 刘建业, 刘昇. 一种基于边缘特征的改进ORB算法[J]. 导航与控制, 2018, 17(2): 18-24.
[3]	彭玮航, 王轲, 刘少鹏, 丁祝顺. 基于区域的卷积神经网络在空对地车辆检测中的应用[J]. 导航与控制, 2017, 16(5): 40-46.
[4]	孙梦玫, 阎璐, 吕明, 缪晓龙. 空中运动目标的识别和跟踪控制[J]. 导航与控制, 2013, 12(4): 6-9.