A reliable multisensor fusion strategy for land vehicle positioning using low-cost sensors

Self Cite

In this paper, we propose a cost-effective localization solution for land vehicles, which can simultaneously adapt to the uncertain noise of inertial sensors and bridge Global Positioning System (GPS) outages. First, three Unscented Kalman filters (UKFs) with different noise covariances are introduced into the framework of Interacting Multiple Model (IMM) algorithm to form the proposed IMM-based UKF, termed as IMM-UKF. The IMM algorithm can provide a soft switching among the three UKFs and therefore adapt to different noise characteristics. Further, two IMM-UKFs are executed in parallel when GPS is available. One fuses the information of low-cost GPS, in-vehicle sensors, and micro electromechanical system (MEMS)-based reduced inertial sensor systems (RISS), while the other fuses only in-vehicle sensors and MEMS-RISS. The differences between the state vectors of the two IMM-UKFs are considered as training data of a Grey Neural Network (GNN) module, which is known for its high prediction accuracy with a limited amount of samples. The GNN module can predict and compensate position errors when GPS signals are blocked. To verify the feasibility and effectiveness of the proposed solution, road-test experiments with various driving scenarios were performed. The experimental results indicate that the proposed solution outperforms all the compared methods.

\begin{matrix} p_{k} = \sin^{- 1} \frac{a_{k}^{x} - {\dot{v}}_{k}^{w h}}{g} \\ r_{k} = - \sin^{- 1} \frac{a_{k}^{y} + ω_{k}^{z} v_{k}^{w h}}{normalg \cos p_{k}} \end{matrix}

where the subscript

k

represents the time step,

p_{k}

and

r_{k}

are pitch and roll angle, respectively.

{\dot{v}}_{k}^{w h}

is the differentiation of

v_{k}^{w h}

g

denotes the acceleration due to gravity.…”

Section: Proposed Imm-ukf Algorithmmentioning

confidence: 99%

m ({\dot{v}}_{k}^{y} + ω_{k}^{z} v_{k}^{w h}) = 2 F_{k}^{s f} + 2 F_{k}^{s r}

where

m

is the mass of the vehicle.

F_{k}^{s f}

and

F_{k}^{s r}

are the front-tire lateral force and the rear-tire lateral force, respectively.…”

Section: Proposed Imm-ukf Algorithmmentioning

confidence: 99%

A Cost-Effective Vehicle Localization Solution Using an Interacting Multiple Model−Unscented Kalman Filters (IMM-UKF) Algorithm and Grey Neural Network

Liu

Chan

2017

Self Cite

“…Using the kinematic relationship between the IMU output and the derivatives of the Euler angles, and assuming that the rotation rate of the Earth is negligible, the longitudinal and lateral motion of the vehicle can be modeled by the equations [ 33 , 34 ]

where w x , w y , and w z are the three angular velocities in the vehicle body frame. v x , v y , and v z are the three linear velocities in the vehicle body frame,

and

are the differentiation of v x and v y , respectively, a x and a y are the longitudinal and lateral accelerations in the vehicle body frame, P and R are pitch and roll angle, respectively, g denotes the acceleration due to gravity.…”

Section: Pitch and Roll Angle Estimationmentioning

confidence: 99%

A Highly Reliable and Cost-Efficient Multi-Sensor System for Land Vehicle Positioning

Liu

et al. 2016

Self Cite

In this paper, we propose a novel positioning solution for land vehicles which is highly reliable and cost-efficient. The proposed positioning system fuses information from the MEMS-based reduced inertial sensor system (RISS) which consists of one vertical gyroscope and two horizontal accelerometers, low-cost GPS, and supplementary sensors and sources. First, pitch and roll angle are accurately estimated based on a vehicle kinematic model. Meanwhile, the negative effect of the uncertain nonlinear drift of MEMS inertial sensors is eliminated by an H∞ filter. Further, a distributed-dual-H∞ filtering (DDHF) mechanism is adopted to address the uncertain nonlinear drift of the MEMS-RISS and make full use of the supplementary sensors and sources. The DDHF is composed of a main H∞ filter (MHF) and an auxiliary H∞ filter (AHF). Finally, a generalized regression neural network (GRNN) module with good approximation capability is specially designed for the MEMS-RISS. A hybrid methodology which combines the GRNN module and the AHF is utilized to compensate for RISS position errors during GPS outages. To verify the effectiveness of the proposed solution, road-test experiments with various scenarios were performed. The experimental results illustrate that the proposed system can achieve accurate and reliable positioning for land vehicles.

“…Providing pedestrians and vehicles with continuous positioning is a challenging but important topic [1,2,3]. In infrastructure-free navigation, it is necessary for mobile platform to provide accurate positioning in seamless indoor/outdoor occasions based on all the required sensors [4,5,6].…”

Section: Introductionmentioning

confidence: 99%

Intelligent Positioning for a Commercial Mobile Platform in Seamless Indoor/Outdoor Scenes based on Multi-sensor Fusion

Wang

Zhang

et al. 2019

Many traffic occasions such as tunnels, subway stations and underground parking require accurate and continuous positioning. Navigation and timing services offered by the Global Navigation Satellite System (GNSS) is the most popular outdoor positioning method, but its signals are vulnerable to interference, leading to a degraded performance or even unavailability. The combination of magnetometer and Inertial Measurement Unit (IMU) is one of the commonly used indoor positioning methods. Within the proposed mobile platform for positioning in seamless indoor and outdoor scenes, the data of magnetometer and IMU are used to update the positioning when the GNSS signals are weak. Because the magnetometer is susceptible to environmental interference, an intelligent method for calculating heading angle by magnetometer is proposed, which can dynamically calculate and correct the heading angle of the mobile platform in a working environment. The results show that the proposed method of calculating heading angle by magnetometer achieved better performance with interference existence. Compared with the uncorrected heading angle, the corrected accuracy results could be improved by 60%, and the effect was more obvious when the interference was stronger. The error of overall positioning trajectory and true trajectory was within 2 m.