轻小型无人机遥感组网飞行的高程安全监测冗余容错算法研究
|
摘要
针对轻小型无人机遥感组网飞行的高程安全要求,本文设计了基于INS/GPS/气压计的多源信息冗余容错测量方案。通过分析轻小型无人机遥感组网应用时复杂多变的工作环境对可靠性与容错能力的要求,采用了联邦滤波算法进行多传感器冗余信息融合。本文分析了联邦滤波结构及算法,通过计算得出了系统容错性好以及滤波精度高的信息分配系数取值原则,并在此基础上提出一种基于故障特征信息熵的Pignistic概率转换容错信息分配方法。该算法可得到清晰准确的故障概率分布,根据此概率分布运用信息熵来确定系统故障概率,进而结合信息分配系数的取值原则得出组合测量系统各个子系统的权重比。通过算例仿真验证了不同信息分配系数主要影响子系统的估计误差和容错性能,而对联邦主滤波器的融合估计误差影响较小,说明了本文的容错信息分配方法能够为各子系统分量提供可靠的分配系数。在旋翼无人机平台上的定高悬停实验证明了该方法能将无人机高程误差减小为传统联邦滤波算法的四分之一,进一步说明了该方法能提高无人机高程安全监测系统的精度及容错性。
To satisfy the altitude safety monitoring for remote sensing networking flight of light and small Unmanned Aerial Vehicles(UAVs), a multi-source information redundancy measurement scheme based on INS/GPS/barometer is designed in this paper. By analyzing the requirement of UAV reliability and fault tolerance in the complex and changeable working environment of remote sensing network application, the federated filtering algorithm is adopted to fuse the redundant information of multi-sensor. In this paper, the structure and algorithm of federated filtering are analyzed, and the principle of selecting information allocation coefficients with good fault-tolerance and high filtering accuracy is obtained. Then a Pignistic probabilistic transformation fault-tolerant information allocation method based on information entropy is proposed. This algorithm can obtain a clear and accurate fault probability distribution, from which the system fault probability is determined by information entropy. Then the weight ratio of each subsystem of the group measurement system is obtained by combining the value principle of information allocation coefficient. Simulation results show that different information allocation coefficients mainly affect the estimation error and fault-tolerant performance of subsystems, but which have little influence on the fusion estimation error of federated main filter. It shows that the fault-tolerant information allocation method in this paper can provide reliable allocation coefficients for each subsystem component. The fixed-altitude hovering experiment on the multi-rotor UAV platform proves that this method can be used to reduce the altitude error of UAV to one quarter of the traditional federated filtering algorithm, which further proves that this method can be used to improve the accuracy and fault-tolerance of the UAV altitude safety monitoring system.
To satisfy the altitude safety monitoring for remote sensing networking flight of light and small Unmanned Aerial Vehicles(UAVs), a multi-source information redundancy measurement scheme based on INS/GPS/barometer is designed in this paper. By analyzing the requirement of UAV reliability and fault tolerance in the complex and changeable working environment of remote sensing network application, the federated filtering algorithm is adopted to fuse the redundant information of multi-sensor. In this paper, the structure and algorithm of federated filtering are analyzed, and the principle of selecting information allocation coefficients with good fault-tolerance and high filtering accuracy is obtained. Then a Pignistic probabilistic transformation fault-tolerant information allocation method based on information entropy is proposed. This algorithm can obtain a clear and accurate fault probability distribution, from which the system fault probability is determined by information entropy. Then the weight ratio of each subsystem of the group measurement system is obtained by combining the value principle of information allocation coefficient. Simulation results show that different information allocation coefficients mainly affect the estimation error and fault-tolerant performance of subsystems, but which have little influence on the fusion estimation error of federated main filter. It shows that the fault-tolerant information allocation method in this paper can provide reliable allocation coefficients for each subsystem component. The fixed-altitude hovering experiment on the multi-rotor UAV platform proves that this method can be used to reduce the altitude error of UAV to one quarter of the traditional federated filtering algorithm, which further proves that this method can be used to improve the accuracy and fault-tolerance of the UAV altitude safety monitoring system.
引文
[1]赵培,王光彦,王之顺,等.消费型无人机在低空航测中的可行性应用研究[J].矿山测量,2018,46(3):20-22.[Zhao P,Wang G Y,Wang Z S,et al.Research on the feasibility application of consumer UAV in low-altitude aerial surveying[J].Mine Survey,2018,46(3):20-22.]
[2]徐志豪.消费级无人机倾斜摄影三维建模可行性研究[J].北京测绘,2018,32(8):897-904.[Xu Z H.Feasibility study on three dimensional modeling of tilting photography for consumer UAV[J].Beijing Surveying and Mapping,2008,32(8):897-904.]
[3]晏磊,吕书强,赵红颖,等.无人机航空遥感系统关键技术研究[J].武汉大学学报(工学版),2004,37(6):67-70.[Yan L,Lv S Q,Zhao H Y,et al.Research on key techniques of aerial remote sensing system for unmanned aerial vehicles[J].Journal of Wuhan University(Engineering Edition),2004,37(6):67-70.]
[4]And D Z,Marta B.From large chemical plant data to fault diagnosis integrated to decentralized fault-tolerant control:Pulp mill process application[J].Industrial&Engineering Chemistry Research,2008,47(4):1201-1220.
[5]鹿强,吴琳,陈昭,等.海上目标多源轨迹数据关联综述[J].地球信息科学学报,2018,20(5):571-581.[Li Q,Wi L,Chen Z,et al.A review of multi-source trajectory data association for marine targets.Journal of Geo-information Science,2018,20(5):571-581.]
[6]Jing L,Wu J,Hou J,et al.Fault-tolerant integrated navigation algorithm of the federal Kalman filter[J].Advances in Future Computer and Control Systems.2012,160(1):621-627.
[7]Gao B,Hu G,Gao S,et al.Multi-sensor optimal data fusion for INS/GNSS/CNS integration based on unscented Kalman filter[J].International Journal of Control Automation&Systems,2018,16(1):129-140.
[8]Yang C,Shi W Z,Chen W.Correlational inference-based adaptive unscented Kalman filter with application in GNSS/IMU integrated navigation[J].GPS Solutions,2018,22(4):100-112.
[9]张亮,高井祥,李增科.信息分配系数的自适应滤波在GPS/DR组合导航中的应用[J].测绘科学技术学报,2015,32(1):18-21.[Zhang L,Gao J X,Li Z K.Application of adaptive filter based on information distribution coefficient to GPS/DR integrated navigation[J].Journal of Geomatics Science and Technology,2015,32(1):18-21.]
[10]马传焱,熊智,许建新,等.一种联邦滤波信息共享分配算法[J].中国惯性技术学报,2013,21(3):347-353.[Ma C Y,Xiong Z,Xu J X,et al.Federated filter information sharing allocation algorithm[J].Journal of Chinese Inertial Technology,2013,21(3):347-353.]
[11]段睿,张小红,朱锋.多源信息融合的组合导航自适应联邦滤波算法[J].系统工程与电子技术,2018,40(2):267-272.[Duan R,Zhang X H,Zhu F.Adaptive federated filter for multi-sources information fusion in integrated navigation system[J].Systems Engineering and Electronics,2018,40(2):267-272.]
[12]李兆铭,杨文革,丁丹,等.多终端仅测速实时定轨的自适应联邦STCKF算法[J].系统工程与电子技术,2017,39(1):177-182.[Li Z M,Yang W G,Ding D,et al.Adaptive federated STCKF algorithm for satellite orbit determination of multiple hand-held terminals[J].System Engineering and Electronics Technology,2017,39(1):177-182.]
[13]邓中亮,尹露,杨磊,等.基于联邦卡尔曼的GPS/基站定位信息融合算法[J].北京邮电大学报,2013,36(6):32-36.[Deng Z L,Yin L,Yang L,et al.Location information fusion algorithm for GPS/Base-Station positioning system based on federated Kalman filter[J].Journal of Beijing University of Posts&Telecommunications,2013,36(6):32-36.]
[14]尹露,邓中亮,席岳.基于可信度的多源定位数据融合方法[J].北京邮电大学学报,2014,37(4):34-38.[Yin L,Deng Z L,Xi Y.Credibility based data fusion algorithm for multi-positioning system[J].Journal of Beijing University of Posts&Telecommunications,2014,37(4):34-38.]
[15]Zhou W,Wang Q,University H E.The INS/GNSS/CNSintegrated navigation system based on vector information distribution[J].Journal of Harbin Institute of Technology,2015,47(4):99-103.
[16]Zhu J Q,Jin W D,Zheng G,et al.High-speed train running gear fault recognition based on information fusion of multi-source[J].Journal of Vibration&Shock,2014,33(21):183-188.
[17]郭承军.多源组合导航系统信息融合关键技术研究[D].成都:电子科技大学,2018.[Guo C J.Key technical research of information fusion for multiple source integrated navigation system[D].Chengdu:University of electronic science and technology,2018.]
[18]袁赣南,袁克非,张红伟,等.联邦滤波器信息分配原则的探讨[J].中国航海,2012,35(3):11-15.[Yuan G N,Yuan KF,Zhang W H,et al.Discussion of information sharing principle for federated filter[J].Navigation of China,2012,35(3):11-15.]
[19]徐晓滨,文成林,王迎昌.基于模糊故障特征信息的随机集度量信息融合诊断方法[J].电子与信息学报,2009,31(7):1635-1640.[Xu X B,Wen C L,Wang Y C.Information fusion algorithm of fault diagnosis based on random set metrics of fuzzy fault features[J].Journal of Electronics&Information Technology,2009,31(7):1635-1640.]
[20]Jiang W,Wu C,Jia J,et al.A probabilistic transformation of basic probability assignment(BPA)in D-S evidence theory[J].Journal of Northwestern Polytechnical University,2013,31(2):295-299.
[21]马丽丽,张芬,陈金广.一种基于Pignistic概率距离的合成公式[J].计算机工程与应用,2015,51(24):61-66.[Ma LL,Zhang F,Chen J G.Synthetic rule of evidence based on Pignistic probability distance[J].Computer Engineering&Applications,2015,51(24):61-66.]
[22]Smets P.Constructing the Pignistic probability function in a context of uncertainty[J].Machine Intelligence&Pattern Recognition,1990,10(1):29-39.
[23]钱华明,朱明,陈亮.信息熵在导航传感器故障诊断中的应用研究[J].系统仿真学报,2010,22(S1):216-219.[Qian H M,Zhu M,Chen L.Information entropy for fault diagnosis of inertia navigation[J].Journal of System Simulation,2010,22(S1):216-219.]
[24]杨可明,魏华锋,刘飞,等.以光谱信息熵改进的N-FINDR高光谱端元提取算法[J].地球信息科学学报,2015,17(8):979-985.[Yang K M,Wei H F,Liu F,et al.Improved N-FINDR algorithm on hyperspectral endmember extraction based on spectral Shannon entropy.2015,17(8):979-985.]
[2]徐志豪.消费级无人机倾斜摄影三维建模可行性研究[J].北京测绘,2018,32(8):897-904.[Xu Z H.Feasibility study on three dimensional modeling of tilting photography for consumer UAV[J].Beijing Surveying and Mapping,2008,32(8):897-904.]
[3]晏磊,吕书强,赵红颖,等.无人机航空遥感系统关键技术研究[J].武汉大学学报(工学版),2004,37(6):67-70.[Yan L,Lv S Q,Zhao H Y,et al.Research on key techniques of aerial remote sensing system for unmanned aerial vehicles[J].Journal of Wuhan University(Engineering Edition),2004,37(6):67-70.]
[4]And D Z,Marta B.From large chemical plant data to fault diagnosis integrated to decentralized fault-tolerant control:Pulp mill process application[J].Industrial&Engineering Chemistry Research,2008,47(4):1201-1220.
[5]鹿强,吴琳,陈昭,等.海上目标多源轨迹数据关联综述[J].地球信息科学学报,2018,20(5):571-581.[Li Q,Wi L,Chen Z,et al.A review of multi-source trajectory data association for marine targets.Journal of Geo-information Science,2018,20(5):571-581.]
[6]Jing L,Wu J,Hou J,et al.Fault-tolerant integrated navigation algorithm of the federal Kalman filter[J].Advances in Future Computer and Control Systems.2012,160(1):621-627.
[7]Gao B,Hu G,Gao S,et al.Multi-sensor optimal data fusion for INS/GNSS/CNS integration based on unscented Kalman filter[J].International Journal of Control Automation&Systems,2018,16(1):129-140.
[8]Yang C,Shi W Z,Chen W.Correlational inference-based adaptive unscented Kalman filter with application in GNSS/IMU integrated navigation[J].GPS Solutions,2018,22(4):100-112.
[9]张亮,高井祥,李增科.信息分配系数的自适应滤波在GPS/DR组合导航中的应用[J].测绘科学技术学报,2015,32(1):18-21.[Zhang L,Gao J X,Li Z K.Application of adaptive filter based on information distribution coefficient to GPS/DR integrated navigation[J].Journal of Geomatics Science and Technology,2015,32(1):18-21.]
[10]马传焱,熊智,许建新,等.一种联邦滤波信息共享分配算法[J].中国惯性技术学报,2013,21(3):347-353.[Ma C Y,Xiong Z,Xu J X,et al.Federated filter information sharing allocation algorithm[J].Journal of Chinese Inertial Technology,2013,21(3):347-353.]
[11]段睿,张小红,朱锋.多源信息融合的组合导航自适应联邦滤波算法[J].系统工程与电子技术,2018,40(2):267-272.[Duan R,Zhang X H,Zhu F.Adaptive federated filter for multi-sources information fusion in integrated navigation system[J].Systems Engineering and Electronics,2018,40(2):267-272.]
[12]李兆铭,杨文革,丁丹,等.多终端仅测速实时定轨的自适应联邦STCKF算法[J].系统工程与电子技术,2017,39(1):177-182.[Li Z M,Yang W G,Ding D,et al.Adaptive federated STCKF algorithm for satellite orbit determination of multiple hand-held terminals[J].System Engineering and Electronics Technology,2017,39(1):177-182.]
[13]邓中亮,尹露,杨磊,等.基于联邦卡尔曼的GPS/基站定位信息融合算法[J].北京邮电大学报,2013,36(6):32-36.[Deng Z L,Yin L,Yang L,et al.Location information fusion algorithm for GPS/Base-Station positioning system based on federated Kalman filter[J].Journal of Beijing University of Posts&Telecommunications,2013,36(6):32-36.]
[14]尹露,邓中亮,席岳.基于可信度的多源定位数据融合方法[J].北京邮电大学学报,2014,37(4):34-38.[Yin L,Deng Z L,Xi Y.Credibility based data fusion algorithm for multi-positioning system[J].Journal of Beijing University of Posts&Telecommunications,2014,37(4):34-38.]
[15]Zhou W,Wang Q,University H E.The INS/GNSS/CNSintegrated navigation system based on vector information distribution[J].Journal of Harbin Institute of Technology,2015,47(4):99-103.
[16]Zhu J Q,Jin W D,Zheng G,et al.High-speed train running gear fault recognition based on information fusion of multi-source[J].Journal of Vibration&Shock,2014,33(21):183-188.
[17]郭承军.多源组合导航系统信息融合关键技术研究[D].成都:电子科技大学,2018.[Guo C J.Key technical research of information fusion for multiple source integrated navigation system[D].Chengdu:University of electronic science and technology,2018.]
[18]袁赣南,袁克非,张红伟,等.联邦滤波器信息分配原则的探讨[J].中国航海,2012,35(3):11-15.[Yuan G N,Yuan KF,Zhang W H,et al.Discussion of information sharing principle for federated filter[J].Navigation of China,2012,35(3):11-15.]
[19]徐晓滨,文成林,王迎昌.基于模糊故障特征信息的随机集度量信息融合诊断方法[J].电子与信息学报,2009,31(7):1635-1640.[Xu X B,Wen C L,Wang Y C.Information fusion algorithm of fault diagnosis based on random set metrics of fuzzy fault features[J].Journal of Electronics&Information Technology,2009,31(7):1635-1640.]
[20]Jiang W,Wu C,Jia J,et al.A probabilistic transformation of basic probability assignment(BPA)in D-S evidence theory[J].Journal of Northwestern Polytechnical University,2013,31(2):295-299.
[21]马丽丽,张芬,陈金广.一种基于Pignistic概率距离的合成公式[J].计算机工程与应用,2015,51(24):61-66.[Ma LL,Zhang F,Chen J G.Synthetic rule of evidence based on Pignistic probability distance[J].Computer Engineering&Applications,2015,51(24):61-66.]
[22]Smets P.Constructing the Pignistic probability function in a context of uncertainty[J].Machine Intelligence&Pattern Recognition,1990,10(1):29-39.
[23]钱华明,朱明,陈亮.信息熵在导航传感器故障诊断中的应用研究[J].系统仿真学报,2010,22(S1):216-219.[Qian H M,Zhu M,Chen L.Information entropy for fault diagnosis of inertia navigation[J].Journal of System Simulation,2010,22(S1):216-219.]
[24]杨可明,魏华锋,刘飞,等.以光谱信息熵改进的N-FINDR高光谱端元提取算法[J].地球信息科学学报,2015,17(8):979-985.[Yang K M,Wei H F,Liu F,et al.Improved N-FINDR algorithm on hyperspectral endmember extraction based on spectral Shannon entropy.2015,17(8):979-985.]