A GPR-PSO incremental regression framework on GPS/INS integration for vehicle localization under urban environment

Land vehicle localization and navigation mainly relies on the Global Positioning System (GPS)/Inertial Navigation System (INS) integration. In this paper, we propose a unified incremental regression framework that enables vehicle localization with high accuracy in urban environment. Within the framework, we propose a nonlinear Gauss Process Regression (GPR) approach to perform vehicle position prediction during GPS outages. By mapping nonlinear data into high-dimensional space by kernel function, the proposed GPR based approach is able to deal with the nonlinearity issue in GPS denied environment. We design a Particle Swarm Optimization (PSO) based algorithm to optimize GPR hyper-parameters, which are tuned with high time efficiency for vehicular position prediction. By real-world road experiments, the proposed method is evaluated against Artificial Neural Network (ANN), Support Vector Machine Regression (SVR) and Partial Least Squares Regression (PLSR). The results reveal that the proposed outperforms the others by 22.8%-65.2% improvement in the positional accuracy.

[1]  Prabir Bhattacharya,et al.  A novel hybrid fusion algorithm to bridge the period of GPS outages using low-cost INS , 2014, Expert Syst. Appl..

[2]  Duy Nguyen-Tuong,et al.  Local Gaussian Process Regression for Real Time Online Model Learning , 2008, NIPS.

[3]  Dong Wang,et al.  Hybrid global navigation satellite systems, differential navigation satellite systems and time of arrival cooperative positioning based on iterative finite difference particle filter , 2015, IET Commun..

[4]  Zhu Xiao,et al.  A Hybrid Approach‐Based Sparse Gaussian Kernel Model for Vehicle State Determination during Outage‐Free and Complete‐Outage GPS Periods , 2016 .

[5]  Shengli Xie,et al.  MixGroup: Accumulative Pseudonym Exchanging for Location Privacy Enhancement in Vehicular Social Networks , 2016, IEEE Transactions on Dependable and Secure Computing.

[6]  Songhwai Oh,et al.  Real-time navigation in crowded dynamic environments using Gaussian process motion control , 2014, 2014 IEEE International Conference on Robotics and Automation (ICRA).

[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]  Dong Wang,et al.  A Nonlinear Framework of Delayed Particle Smoothing Method for Vehicle Localization under Non-Gaussian Environment , 2016, Sensors.

[9]  Gong Zhang,et al.  GPS/INS Integrated Navigation Based on UKF and Simulated Annealing Optimized SVM , 2013, 2013 IEEE 78th Vehicular Technology Conference (VTC Fall).

[10]  Dong Wang,et al.  A Gaussian mixture framework for incremental nonparametric regression with topology learning neural networks , 2016, Neurocomputing.

[11]  Ming Tang,et al.  Hierarchical and Networked Vehicle Surveillance in ITS: A Survey , 2015, IEEE transactions on intelligent transportation systems (Print).

[12]  S. Wold,et al.  The multivariate calibration problem in chemistry solved by the PLS method , 1983 .

[13]  Hong Wang,et al.  A low-cost INS/GPS integration methodology based on random forest regression , 2013, Expert Syst. Appl..

[14]  Christopher K. I. Williams,et al.  Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning) , 2005 .

[15]  Yue Shi,et al.  A modified particle swarm optimizer , 1998, 1998 IEEE International Conference on Evolutionary Computation Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98TH8360).

[16]  Fakhri Karray,et al.  Multisensor data fusion: A review of the state-of-the-art , 2013, Inf. Fusion.

[17]  Dong Wang,et al.  Online-SVR for vehicular position prediction during GPS outages using low-cost INS , 2015, 2015 IEEE 26th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC).