小型无人机多传感器组合导航技术研究

小型无人机以其机动灵活性和实用经济性，使得应用场景在不断扩展，对飞 行可靠性的要求也随之更高，这就要求无人机具备拥有高精度、满足更多场景的 导航定位能力，单一的导航系统已经难以胜任这些导航需要。组合导航技术通过 利用不同导航系统形成优势互补，提升无人机导航精度和稳定性。基于此，本文 在捷联惯性导航系统的基础上、结合全球导航卫星系统和光流导航进行针对无人 机的组合导航系统研究工作，主要包含以下方面内容： 文章首先对捷联惯性导航的原理进行研究，针对初始对准阶段微惯性传感器 无法自主寻北问题，结合磁强计进行航向初始对准，在此基础上研究了捷联惯性 导航机械编排算法和误差方程，针对小型无人机导航特点，对误差模型进行合理 简化，并利用 Allan 方差分析对惯性器件的随机误差进行分析和建模。 针对组合导航系统，文章基于松组合方式，利用间接卡尔曼滤波反馈校正的 方法，构建起卫星/惯导组合导航模型，结合不同导航方式的各自优点，实现导航 系统的数据融合。同时，针对无卫星导航信号的导航场景，文章引入光流传感器， 通过光流解算获得速度和位置信息，结合惯性导航系统建立了光流/惯性组合导航 模型，对传统的组合导航方案进行了补充，为无人机应用范围的扩展，提供了新的 导航方案。 针对卫星信号在复杂场景下受外界环境影响，使得系统噪声发生改变，造成 组合导航系统精度下降的问题，本文提出了一种自适应卡尔曼滤波方法。文章通 过主成分分析获取卫星导航定位特征参数中和导航定位精度相关的主要特征参数， 利用实时获得的接收机卫星导航参数结合模糊推理算法，对组合导航中的传统卡 尔曼滤波进行改进，实现根据卫星导航信号质量对量测噪声协方差阵的自适应，有 效提高了组合导航系统滤波精度和稳定性。 基于搭建的硬件平台，通过软硬件联调联试，对设计的组合导航系统及其相 关算法进行实验测试，验证了本文设计的组合导航系统的有效性。

[1] 吴和龙. 多旋翼无人机的低成本 Inertial/GNSS/Vision 组合导航关键技术研究[D]. 中国科学院大学 (中国科学院长春光学精密机械与物理研究所), 2020.
[2] GROVES P D. Principles of GNSS, inertial, and multisensor integrated navigation systems,[Book review][J]. IEEE Aerospace and Electronic Systems Magazine, 2015, 30(2): 26-27.
[3] 马福民, 王惠. 微系统技术现状及发展综述[J]. 电子元件与材料, 2019, 38(6): 8.
[4] 张珂. 基于光流和惯性导航的四旋翼无人机定位方法研究[J]. 文摘版：自然科学, 2016,000(004): P.311-311.
[5] 秦永元. 惯性导航. 第 2 版[M]. 惯性导航. 第 2 版, 2014.
[6] 李征航, 黄劲松. GPS 测量与数据处理. 第 2 版[M]. GPS 测量与数据处理. 第 2 版, 2010.
[7] 任珊珊. 多旋翼无人机 INS/GNSS 组合导航方法研究[D]. 北京理工大学, 2018.
[8] 柴大帅. 多星座 GNSS/INS 组合导航理论与方法研究[D]. 中国矿业大学, 2020.
[9] ZHANG T, NIU X, BAN Y, et al. Modeling and Development of INS-Aided PLLs in aGNSS/INS Deeply-Coupled Hardware Prototype for Dynamic Applications[J]. Sensors, 2015,15(1): 733-759.
[10] 叶飞. 城市环境下多频宽巷 GNSS/INS 紧组合定位关键技术研究[D]. 东南大学, 2021.
[11] 温雅丽. 面向小型旋翼无人机的低成本组合导航算法研究[D]. 浙江大学, 2016.
[12] LE N H, NGUYEN V H. Loosely coupled GPS/INS integration with Kalman filtering for landvehicle applications[C]//International Conference on Control. 2013.
[13] YANG Y, GAO W. An optimal adaptive Kalman filter[J]. Journal of geodesy, 2006(4): 80.
[14] RINARA, WOO, EUN-JU, et al. A Fuzzy-Innovation-Based Adaptive Kalman Filterfor Enhanced Vehicle Positioning in DenseUrban Environments.[J]. Sensors, 2019.
[15] KANG X, HE G, LI X. A SINS/BDS Integrated Navigation Method Based on ClassifiedWeighted Adaptive Filtering[J]. Mathematical Problems in Engineering, 2019, 2019.
[16] NING Y, WANG J, HAN H, et al. An Optimal Radial Basis Function Neural Network EnhancedAdaptive Robust Kalman Filter for GNSS/INS Integrated Systems in Complex Urban Areas[J].Sensors, 2018, 18(9).
[17] HAO Y, XU A, SUI X, et al. A Modified Extended Kalman Filter for a Two-Antenna GPS/INSVehicular Navigation System[J/OL]. Sensors, 2018, 18(11). https://www.mdpi.com/1424-8220/18/11/3809. DOI: 10.3390/s18113809.
[18] XIONG L, XIA X, LU Y, et al. IMU-based Automated Vehicle Body Sideslip Angle and AttitudeEstimation Aided by GNSS using Parallel Adaptive Kalman Filters[J]. IEEE Transactions onVehicular Technology, 2020, PP(99): 1-1.
[19] 储诚涛. 基于深度学习的 GNSS/SINS 组合定位误差抑制技术研究[D]. 东南大学, 2021.
[20] HUANG X, CHEN G, LIU Z. Sage Husa Adaptive Integrated Navigation Algorithm Basedon Variable Fading Factor[C/OL]//2020 International Conference on Computer Network, Electronic and Automation (ICCNEA). 2020: 378-383. DOI: 10.1109/ICCNEA50255.2020.00084.
[21] LIU R, LIU F, LIU C, et al. Modified Sage-Husa Adaptive Kalman Filter-Based SINS/DVLIntegrated Navigation System for AUV[J]. Journal of Sensors, 2021.
[22] SUN J, WANG F. An effective LS-SVM/AKF aided SINS/DVL integrated navigation systemfor underwater vehicles[J]. Peer-to-Peer Networking and Applications, 2022, 15(3): 1437-1451.
[23] 李帅. GNSS/SINS 组合导航系统的研究与实现[D]. 桂林电子科技大学, 2021.
[24] YUE S, CONG L, QIN H, et al. A Robust Fusion Methodology for MEMS-based Land VehicleNavigation in GNSS-Challenged Environments[J]. IEEE Access, 2020, PP(99): 1-1.
[25] BROSSARD M, BARRAU A, BONNABEL S. AI-IMU Dead-Reckoning[J]. IEEE Transactionson Intelligent Vehicles, 2020, PP(99): 1-1.
[26] 陶毅峰. GRU 循环神经网络辅助的 GNSS/INS 组合导航算法研究[D]. 武汉大学, 2021.
[27] WEI X, LI J, FENG K, et al. A Mixed Optimization Method Based on Adaptive Kalman Filterand Wavelet Neural Network for INS/GPS During GPS Outages[J]. IEEE Access, 2021, PP(99): 1-1.
[28] JARADAT M A K, ABDEL-HAFEZ M F. Non-Linear Autoregressive Delay-Dependent INS/GPS Navigation System Using Neural Networks[J/OL]. IEEE Sensors Journal, 2017, 17(4):1105-1115. DOI: 10.1109/JSEN.2016.2642040.
[29] TAN X, WANG J, JIN S, et al. GA-SVR and Pseudo-position-aided GPS/INS Integration duringGPS Outage[J/OL]. Journal of Navigation, 2015, 68: 678-696. https://api.semanticscholar.org/CorpusID:26131884.
[30] ADUSUMILLI S, BHATT D, WANG H, et al. A novel hybrid approach utilizing principalcomponent regression and random forest regression to bridge the period of GPS outages[J/OL].Neurocomput., 2015, 166(C): 185–192. https://doi.org/10.1016/j.neucom.2015.03.080.
[31] DELUCIA P C R, JONES K S. Gibson and Crooks (1938): Vision and validation[J]. AmericanJournal of Psychology, 2017, 130(4): 413-429.
[32] ILLEPERUMA G D, SONNADARA U J. An autonomous robot navigation system based onoptical flow[C/OL]//2011 6th International Conference on Industrial and Information Systems.2011: 489-492. DOI: 10.1109/ICIINFS.2011.6038119.
[33] PASTOR-MORENO D, SHIN H S, WALDOCK A. Optical flow localisation and appearance mapping (OFLAAM) for long-term navigation[C]//2015 International Conference on Unmanned Aircraft Systems (ICUAS). IEEE, 2015: 980-988.
[34] ZHANG H, XIAO L, XU G. A novel tracking method based on improved FAST corner detectionand pyramid LK optical flow[C]//2020 Chinese control and decision conference (CCDC). IEEE,2020: 1871-1876.
[35] 魏家琦. 基于改进光流法的自主测速与定位技术[D]. 中北大学, 2022.
[36] DOSOVITSKIY A, FISCHER P, ILG E, et al. Flownet: Learning optical flow with convolutional networks[C]//Proceedings of the IEEE international conference on computer vision.2015: 2758-2766.
[37] 化雪荟, 陈大力. INS/光流/磁强计组合导航在小型无人机中的应用[J]. 电子器件, 2017, 40(6): 7.
[38] MA Z, LI H, GU Y, et al. Flight and Hover Control System Design for a Mini-quadrotor Basedon Multi-sensors[J]. International Journal of Control Automation and Systems, 2019.
[39] 王亭亭, 蔡志浩, 王英勋. 无人机室内视觉/惯导组合导航方法[M]. 北京航空航天大学学报, 2018.
[40] MENG F, YANG D. Research of UAV Location Control System Based on SINS, GPS andOptical Flow[C/OL]//2020 IEEE International Conference on Information Technology,Big Dataand Artificial Intelligence (ICIBA): volume 1. 2020: 495-498. DOI: 10.1109/ICIBA50161.2020.9276977.
[41] 严恭敏，翁浚. 捷联惯导算法与组合导航原理[M]. 西北工业大学出版社, 2019.
[42] 何佳敏, 齐红丽, 雷华明, 等. 融合转向差校正的磁强计地磁补偿硬件实现研究[J]. 仪表技术与传感器, 2021(9): 6.
[43] SAVAGE P G. Strapdown Analytics[Z]. 2000.
[44] JOHNH.MATHEWS), FINK K. 数值方法:MATLAB 版[M]. 数值方法:MATLAB 版, 2005.
[45] SHIN E H. Estimation techniques for low-cost inertial navigation[M]. University of Calgary,Department of Geomatics Engineering, 2005.
[46] WU Z, WANG W. INS/magnetometer integrated positioning based on neural network for bridging long-time GPS outages[J]. GPS Solutions, 2019, 23(3).
[47] ALLAN D W. Statistics of Atomic Frequency Standards[J]. Proceedings of the IEEE, 1966, 54(2): 221-230.
[48] IEEE B E. IEEE Standard Specification Format Guide and Test Procedure for Single-AxisInterferometric Fiber Optic Gyros[J]. IEEE, 1998.
[49] BAI Y, XIAOYIJIN, XUEBOSU, et al. Adaptive filtering for MEMS gyroscope with dynamicnoise model[J]. ISA Transactions, 2020, 101(1).
[50] 宁津生, 姚宜斌, 张小红. 全球导航卫星系统发展综述[J/OL]. 导航定位学报, 2013, 1(01):3-8. DOI: 10.16547/j.cnki.10-1096.2013.01.005.
[51] GIBSON J J. The perception of the visual world.[M]. Houghton Mifflin, 1950.
[52] BEAUCHEMIN S S, BARRON J L. The computation of optical flow[J]. ACM computingsurveys (CSUR), 1995, 27(3): 433-466.
[53] 黄赞, 张宪民. 改进的基于光流的鲁棒多尺度运动估计算法[J]. 华南理工大学学报：自然科学版, 2009(11): 5.
[54] WELCH G, BISHOP G. An introduction to the Kalman filter (Technical Report No. TR 95-041)[Z]. 1995.
[55] 黄莺. 基于 CKF 的 SINS/DVL 组合导航系统设计与仿真[D]. 哈尔滨工程大学, 2013.
[56] GUMILAR I, MAHDIYANTO R A, BRAMANTO B, et al. Study of GNSS Multi-ConstellationPerformance in Single Point Positioning and Differential Positioning in Indonesia[J]. Journalof Aeronautics, Astronautics and Aviation, A, 2022(2): 54.
[57] KLIR G J, YUAN B. Fuzzy Sets and Fuzzy Logic[Z]. 1995.