數位式移動測繪系統(DMMS)是一種整合數位影像與直接定位定向的技術。而直接定位定向用於直接計算影像於拍攝時間點其影像製造平台的空間軌跡之時變位置與時變姿態參數。現今最通用之技術為使用全球定位系統(GPS)作為位置感測的主要元件同時使用慣性導航系統(INS)作為姿態感測的主要元件。

卡曼濾波器(KF)被視為即時性INS/GPS整合系統的最佳預估工具。這是一組數學方程式可提供具有高計算效率及可預估狀態的連續過程。然而，KF已經被證實具有多項限制，而為了消除其限制與改進此演算法的效能，人工類神經網路(ANN)與KF之混合式智慧型架構已被提出成為新的INS/GPS整合系統之取代技術。另外，與即時性模式相比的後處理模式，其具有能使用整條軌跡資料去估計軌跡的優點，故如RTS後向平滑之最佳化平滑器可在後處理模式下有效提升精度等級。如今商業化的移動式測繪系統即是使用RTS後向平滑演算法來提供高精度的位置與姿態解以運行直接定位定向之解算程序。

本研究開發與分析整合微機電系統(MEMS)之慣性量測元件(IMU)與GPS接收機之發展。所提出的架構運用自成長式類神經網路(階層關聯式神經網路，CCN)以克服以KF為基礎的傳統技術其所存在之限制。使用CCN於INS/GPS整合位置與姿態系統(POS)，其相較於多層前饋式網路(MFNN)更具有彈性的位相關係、可從新軌跡所獲得之知識自行成長神經元已吸納新資訊及提供較穩定的誤差補償。從初步實驗結果得知，CCN_POS與MFNN_POS為基礎的傳統KF技術用於INS/GPS整合系統，其在位置與姿態精度方面比起單使用KF與RTS具有較佳之精度與效率，且CCN_POS適用吸收多維的輸入資訊以正確評估誤差的行為，其再成長的學習架構可有效記憶新的資訊，以預測新環境下的誤差行為。

The technique that integrates digital imaging with direct-georeferencing is known as a digital mobile mapping system (DMMS). Direct-georeferencing is a new alternative to directly decide the time-variable position and orientation parameters for orienting the images in space, trajectory and attitude of the imaging platform. The most common technologies used for this purpose today are using Global Positioning System (GPS) as a major position sensor, while the Inertial Navigation System (INS) is the major orientation sensor.

Kalman Filter (KF) has been considered as the optimal estimation tool for real-time INS/GPS. The KF is a set of mathematical equations that provides an efficient computational means to estimate the state of a process. However, KF has been approved to have several limitations. In order to eliminate these limitations and improve the performance of an INS/GPS integrated system, and intelligent and hybrid scheme consists of an Artificial Neural Networks (ANN) and KF has been applied to develop alternative INS/GPS integration schemes. Post-mission processing, when compared to real-time filtering, has the advantage that data of the whole mission can be used to estimate the trajectory. The optimal smoothing method such as Rauch-Tung-Striebel (RTS) backward smoother can be applied in post mission mode. Most of the commercial mobile mapping systems use an optimal smoothing algorithms to provide accurate positioning and orientation for direct geo-referencing.

This study exploits and analyzes the idea of developing an alternative data fusion scheme that integrates the outputs of a low cost Micro-Electro-Mechanical Systems (MEMS) Inertial Measurements Units (IMU) and GPS receivers. The proposed scheme is implemented using a constructive neural network (Cascade-Correlation Network, CCN) to overcome the limitations of conventional techniques that are based on the Kalman filter. The CCN applied in this research has the advantage of having a flexible topology, growing neurons from new data sets in different trajectory and stable outputs if compared to the recently utilized Multi-layer Feed-forward Neural Networks (MFNN) for INS/GPS integrated Positioning and Orientation Systems (POS). The preliminary results presented in this article illustrate the effectiveness of proposed CCN over both MFNN-based and Kalman filtering based techniques for INS/GPS integration. In addition, the finding of this research indicates that the proposed ANN-KF and ANN-RTS hybrid INS/GPS integration schemes can provide superior performance in terms of positioning and orientation accuracies comparing to Kalman filter and RTS smoother, respectively. Ultimately, the proposed CCN based schemes are more reliable to predict correct error behavior by absorbing more input vectors.

1.Alpaydin, E. (1991): GAL, Neural Networks that grow when they learn and shrink when they forget, International Computer Science Institute, TR 91-032.
2.Anderson, D. and McNeill, R. (1992): Artificial Neural Network Technology, Technical Reports, ITT Industries.
3.Bishop, C.M. (1995): Neural Networks for Pattern Recognition, Oxford University Press, New York, NY, USA.
4.Brown, R.G. and Hwang, P.Y.C. (1992): Introduction to random signals, John Wiley and Sons, San Francisco, New York, USA.
5.Chang, H.W. and Li, C.Y. (2008): An ANN Embedded POS Algorithm for a low cost MEMS INS/GPS Integrated System, GNSS 2008, ION, Savannah, GA, USA.
6.Chiang, K.W. (2004): INS/GPS Integration Using Neural Networks for Land Vehicular Navigation Applications, Department of Geomatics Engineering, the University of Calgary, Calgary, Canada.
7.Chiang, K.W., Nassar, S. and El-Sheimy N. (2006): A Constructive and Autonomous Integration Scheme of Low-Cost GPS/MEMS IMU for Land Vehicular Navigation Application, IEEE/ION Position, Location, And Navigation Symposium.
8.El-Sheimy, N. (2002): Introduction to Inertial Navigation. ENGO 699.71 Lecture Notes, Geomatics Department, University of Calgary,Calgary, Canada.
9.El-Sheimy, N. (1996): The development of VISAT - A mobile survey system for gis applications. Ph.D. thesis, UCGE Report No. 20101, Department of Geomatics Engineering, the University of Calgary, Calgary, Canada.
10.Fahlman, S.E. and Lebiere, C. (1990): The cascade learning Architecture, In D.S. Touretzky (Ed.), Advances in Neural Information Processing System 2, Denver, CO, Morgan Kaufmann Puvlishers, pp.198-209.
11.Farrell, J.A. and Barth, M. (1998): The Global Positioning System & Inertial Navigation, McGraw-Hill.
12.Gelb, A. (1974): Applied Optimal Estimation, The M.I.T. Press, Cambridge, MA, USA.
13.Goad, C.C. (1991): The Ohio State University Mapping System: The Positioning Component. In Proceedings of the 47th annual Meeting, Williamsburg, VA, June 10–12, 1991, pp. 121–124.
14.Goodall, C., El-Sheimy, N. and Chiang, K.W. (2005): The Development of a GPS/MEMS INS Integrated System Utilizing a Hybrid Processing Architecture. Institute of Navigation GNSS 2005 proceedings, Long Beach, California.
15.Haykin, S. (1991): Neural Networks, A Comprehensive Foundation, second Edition, Prentice-Hall, New Jersey.
16.Hofmann-Wellenhof, B., Lichtenegger, H. and Colins, J.(1997): GPS Theory and Practice, Springer-Verlag Qien New Tork, Austria.
17.Honavar, V. and Uhr, L. (1994): Artificial Intelligence and Neural Networks,Steps Toward Principled Integration, Boston: Academic Press.
18.Kalman, R.E.(1960): A New Approach to Linear Filtering and Prediction Problem, ASME Journal of Basic Engineering, Vol.82, pp.34-35.
19.M. Rokibul Alam, M. Shamsul Huda, M. Mirza Hafiz AI-asad and M. Mostafa Zaman (2004). A Comparison of Different Constructive Algorithms to Design Neural Network, in Proc 2nd International Conference on Electrical and Computer Engineering (ICECE-2004), Dhaka, Bangladesh.
20.Krakiwsky, E.J. (1991): GPS and Vehicle Location and Navigation, Innovation Column, GPS World, Vol. 2, No. 5, Advanstar Communications, Eugene, Oregon, USA.
21.Lin, C.Y. (2008): The Development of Hybrid INS/GPS integration Schemes Using Artificial Neural Network, Master’s thesis, Department of Geomatics, National Cheng Kung University.
22.Lockheed Martin Corporation, retrieved (2005), from http://www.lockheedmartin.com/wms/findPage,do?dsp=fec&ci=17164&rsbci=0&fti=112&ti=0&sc=400.
23.Lu, G.(1995): Development of a GPS Multi-Antenna System for Attitude Determination, Ph.D. dissertation, Department of Geomatics Engineering, The University of Calgary, Calgary, Canada.
24.Meditch, J.S. (1969): Stochastic Optimal Linear Estimation and Control, McGraw-Hill.
25.Mendel, J.M. and McLaren, R.W. (1970): Reinforcement-learning control and pattern recognition systems, in Adaptive, Learning, and Pattern Recognition Systems, Theory and Application, vol. 66. J.M. Mendel and K.S.Fu, eds., pp.287-318, New York: Academic Press.
26.Misra, P. and Enge, P. (2001): Global Positioning System, Signals, Measurements and Performance.
27.Sulej, R. (2008): NetMaker software, from http://www.ire.pw.edu.pl/~rsulej/NetMaker.
28.Park, M. and Gao, Y. (2008): Error and Performance Analysis of MEMS-based Inertial Sensors with a Low-cost GPS Receiver, Sensors, 8, pp.2240-2261.
29.Phatak, D.S. and Koren, I. (1994): Connectivity and Performance Tradeoffs in the Cascade Correlation Learning Architecture, IEEE Trans, Neural Nets, vol. 5, pp.930-935.
30.Phillip, R. E. and Schmidt, G.T. (1996). GPS/INS Integration. AGARD Lecture Series MSP LS 207 on System Implications and Innovative Application of Satellite Navigation, NATO, Paris.
31.Potter, M.A. (1992): A Genetic Cascade-Correlation Learning Algorithm. In L. D. Whitley and J. D. Schaffer (Eds.), COGANN-92 International Workshop on Combinations of Genetic Algorithms and Neural Networks, IEEE Computer Society Press, pp. 123–133.
32.Prechelt, L. (1997): Investigation of the CasCor Family of Learning Algorithms, Neural Networks, vol. 10, no. 5.
33.Rogers, R.M. (2003): Applied Mathematics in Integrated Navigation Systems, Second Edition, American Institute of Aeronautics and Astonautics, Inc.
34.Schwarz, K.P. (1999): Fundamentals of Geodesy, Lecture Notes ENGO 421, Dept. of Geomatics Eng., The University of Calgary, Calgary, Canada.
35.Schwarz, K.P. and Wei, M.(1999): INS/GPS Integration for Geodetic Applications, Lecture Notes ENGO 623, Department of Geomatics Engineering, The University of Calgary, Calgary, Canada.
36.Shin E.H. (2001): Accuracy Improvement of Low Cost INS/GPS for Land Applicatins.
37.Shin, E.H. and El-Sheimy, N. (2005): Navigation Filter Design for Pipeline Pigging, J. Navig. 2005, 58, 283-295.
38.Tao, V. and Li, J. (2007): Advance in Mobile Mapping Technology, International Society for Photogrammetry and Remote Sensing (ISPRS) Book Series, Taylor and Francis Group, London, UK.
39.Titterton, D.H. and Weston, J.L. (1997): Strapdown Inertial Navigation Technology, Peter Peregrinus Ltd.