Reducing complexity of GPS/INS integration scheme through neural networks

A vehicle-mounted GPS receiver used for localization can suffer from signal blockage. To remedy to this problem, GPS/ INS integration can be considered as a solution in some cases. However, in the case of urban areas where there are severe multipath conditions, the performance degrades considerably. The last decade have seen many proposals of techniques aiming at improving the accuracy of GPS positions. The complexity of these techniques increases with the increase of the accuracy required. These techniques usually combine Kalman Filters (KF) with neural network or fuzzy logic... etc. In this paper, we propose a new technique based solely on neural network, which offers a better performance while presenting a lower complexity. The idea is to use a neural network, which emulates the behavior of a given estimator in order to replace it. We present simulations results, which validate the performance and the robustness of our proposed scheme in various conditions.

[1]  Jang Gyu Lee,et al.  Adaptive Two-Stage Extended Kalman Filter for a Fault-Tolerant INS-GPS Loosely Coupled System , 2009, IEEE Transactions on Aerospace and Electronic Systems.

[2]  Raja Sengupta,et al.  Kalman Filter-Based Integration of DGPS and Vehicle Sensors for Localization , 2005, IEEE Transactions on Control Systems Technology.

[3]  D. Boukhetala,et al.  Decentralized neural network controller for trajectory tracking of robot manipulators , 2001 .

[4]  Wolfgang Effelsberg,et al.  Dead-Reckoning for Position-Based Forwarding on Highways , 2006 .

[5]  Otman A. Basir,et al.  Toward increasing the localization accuracy of vehicles in VANET , 2009, 2009 IEEE International Conference on Vehicular Electronics and Safety (ICVES).

[6]  Shahrokh Valaee,et al.  Vehicle Localization in Vehicular Networks , 2006, IEEE Vehicular Technology Conference.

[7]  Naser El-Sheimy,et al.  The Utilization of Artificial Neural Networks for Multisensor System Integration in Navigation and Positioning Instruments , 2006, IEEE Transactions on Instrumentation and Measurement.

[8]  C. Chui,et al.  Approximation by ridge functions and neural networks with one hidden layer , 1992 .

[9]  V. Vaidehi,et al.  Integration of INS and GPS using radial basis function neural networks for vehicular navigation , 2010, 2010 11th International Conference on Control Automation Robotics & Vision.

[10]  T.E. Melgard,et al.  GPS signal availability in an urban area-receiver performance analysis , 1994, Proceedings of 1994 IEEE Position, Location and Navigation Symposium - PLANS'94.

[11]  Kai-Wei Chiang,et al.  An intelligent navigator for seamless INS/GPS integrated land vehicle navigation applications , 2008, Appl. Soft Comput..

[12]  M. Nguyen-H,et al.  Improving GPS/INS Integration through Neural Networks , 2010, ArXiv.

[13]  G.D. Buckner,et al.  Active vehicle suspension control using intelligent feedback linearization , 2000, Proceedings of the 2000 American Control Conference. ACC (IEEE Cat. No.00CH36334).

[14]  John B. Moore,et al.  Direct Kalman filtering approach for GPS/INS integration , 2002 .

[15]  Otman A. Basir,et al.  Intervehicle-Communication-Assisted Localization , 2010, IEEE Transactions on Intelligent Transportation Systems.

[16]  A. Guessoum,et al.  Decentralised neural control of active suspensions , 2005 .

[17]  V. Piuri,et al.  Neural technologies for increasing the GPS position accuracy , 2008, 2008 IEEE International Conference on Computational Intelligence for Measurement Systems and Applications.

[18]  F. Gagnon,et al.  Fuzzy corrections in a GPS/INS hybrid navigation system , 2004, IEEE Transactions on Aerospace and Electronic Systems.