PSO-LSSVR Assisted GPS/INS Positioning in Occlusion Region

Abstract

Satellite signals are easily lost in complex observation environments and high dynamic motion states, and the position and posture errors of pure inertial navigation quickly diverges with time. This paper therefore proposes a scheme of occlusion region navigation based on least squares support vector regression (LSSVR), and particle swarm optimization (PSO), used to seek the global optimal parameters. Firstly, the scheme uses the incremental output of GPS (Global Positioning System) and Inertial Navigation System (INS) when the observation is normal as the training output and the training input sample, and then uses PSO to optimize the regression parameters of LSSVR. When the satellite signal is unavailable, the trained mapping model is used to predict the GPS pseudo position. Secondly, the observed anomaly is detected by the test statistic in the integrated navigation solution filtering estimation, and the exponential fading adaptive factor is introduced to suppress the influence of the abnormal pseudo observation value. The results indicate that the algorithm can predict the higher precision GPS position increment, and can effectively judge some abnormal observations that may occur in the predicted value, and adjust the observed noise covariance to suppress the anomaly observation, which can effectively improve the continuity and reliability of the integrated navigation system in the occlusion region.

Placement accuracy algorithm for smart street lights

“…Nevertheless, the GPS module itself has its own problem in terms of accuracy due to multipath loss, satellite network and atmospheric conditions. Works on improving GPS accuracy, such as by using differential GPS [19], assisted GPS [20,21], Kalman filter [22,23] and precise point positioning technique [24], are ongoing. However, the focus of these works is not on improving the accuracy at the device or system level but on improving available GPS data and performing postprocessing by using a geometrical method.…”

A comparison study between different kernel functions in the least square support vector regression model for penicillin fermentation process

“…Work from [8] incorporated the least-square method with SVR, later termed Least Square Support Vector Regression (LSSVR) has greatly improved modelling efficiency as the complex optimisation is replaced by solving series of linear equations, making LSSVR an ideal algorithm for large scale regression problem while maintaining the great performance of SVR when data are limited. The core idea behind LSSVR is to translate nonlinear samples from lowdimensional space into high-dimensional feature space by employing the kernel function, allowing the nonlinear samples to be partitioned linearly in this space to fulfil the fitting prediction requirement [9].…”

“…Due to the lower computational complexity of the optimisation process in LSSVR as compared to the SVR model, therefore, it has been broadly used in the prediction of quality variables in nonlinear processes. For instance, [9] developed an LSSVR model with a Gaussian kernel function that can accurately assess some anomalous observations that may take place in the estimated value and forecast GPS (Global Positioning System) signals with improved precision. Besides that, [10] managed to design a smart LSSVR model by adopting the Gaussian radial basis function (RBF) kernel function to address batch operations' time-varying, multiphase and nonlinear features.…”

An unscented Kalman filter based velocity estimation method for articulated steering vehicle using a novel dynamic model

“…Global positioning system (GPS) can directly measure the velocity, but it may lose satellite signal frequently in mountainous areas, mining areas, and complex environments with buildings. 10 On the contrary, the inertial measurement unit (IMU) does not need to receive the satellite signal, but the shortage is that it may cause significant accumulated errors 3 because ASV is prone to vibration and oscillation during operation. The state estimation technology is a good solution to overcome these problems as it is based on the fusion of multi-sensor data and the vehicle model.…”