Limits...
Improving Localization Accuracy: Successive Measurements Error Modeling.

Ali NA, Abu-Elkheir M - Sensors (Basel) (2015)

Bottom Line: Vehicle self-localization is an essential requirement for many of the safety applications envisioned for vehicular networks.We prove the existence of correlation between successive measurements in the two datasets, and show that the time correlation between measurements can have a value up to four minutes.Our model can assist in providing better modeling of positioning errors and can be used as a prediction tool to improve the performance of classical localization algorithms such as the Kalman filter.

View Article: PubMed Central - PubMed

Affiliation: College of Information Technology, United Arab Emirates University, Al-Ain 15551, Abu Dhabi. najah@uaeu.ac.ae.

ABSTRACT
Vehicle self-localization is an essential requirement for many of the safety applications envisioned for vehicular networks. The mathematical models used in current vehicular localization schemes focus on modeling the localization error itself, and overlook the potential correlation between successive localization measurement errors. In this paper, we first investigate the existence of correlation between successive positioning measurements, and then incorporate this correlation into the modeling positioning error. We use the Yule Walker equations to determine the degree of correlation between a vehicle's future position and its past positions, and then propose a -order Gauss-Markov model to predict the future position of a vehicle from its past  positions. We investigate the existence of correlation for two datasets representing the mobility traces of two vehicles over a period of time. We prove the existence of correlation between successive measurements in the two datasets, and show that the time correlation between measurements can have a value up to four minutes. Through simulations, we validate the robustness of our model and show that it is possible to use the first-order Gauss-Markov model, which has the least complexity, and still maintain an accurate estimation of a vehicle's future location over time using only its current position. Our model can assist in providing better modeling of positioning errors and can be used as a prediction tool to improve the performance of classical localization algorithms such as the Kalman filter.

No MeSH data available.


Zoom in on the longitudinal positioning error produced by the Autoregression-Kalman filter for vehicle.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4541843&req=5

sensors-15-15540-f014: Zoom in on the longitudinal positioning error produced by the Autoregression-Kalman filter for vehicle.

Mentions: We can see that incorporating our error model into the Kalman filter outperforms the results produced by the Kalman filter with zero Gaussian error, which we call the standard Kalman filter for convenience. Similar accuracy gains are obtained when the error model is integrated into the Kalman filter and applied to the location measurements of vehiclewhose location information is used for validation. We can also observe that there is no observable gain in accuracy when using a higher Gauss–Markov order, sinceprovides almost identical results to. The average measurement error for the standard Kalman filter applied to the location measurements of vehicleis 56 m and 29 m for x and y, respectively, compared to 0.25 m produced by the Kalman filter that is integrated with the autoregression model of order 1. The Kalman filter that was integrated with the autoregression model of order 4 has an average measurement error of 0.24 m, which is very close to that produced by the autoregression model of order 1. This can be further illustrated in Figure 14 and Figure 15, which provide a more detailed view of the behavior of the model for. The fluctuations in error measurements are due to the constant variability in the vehicles’ position measurements, as was found upon further inspection of the original time-series for longitudinal and latitudinal location measurements of both vehicles.


Improving Localization Accuracy: Successive Measurements Error Modeling.

Ali NA, Abu-Elkheir M - Sensors (Basel) (2015)

Zoom in on the longitudinal positioning error produced by the Autoregression-Kalman filter for vehicle.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4541843&req=5

sensors-15-15540-f014: Zoom in on the longitudinal positioning error produced by the Autoregression-Kalman filter for vehicle.
Mentions: We can see that incorporating our error model into the Kalman filter outperforms the results produced by the Kalman filter with zero Gaussian error, which we call the standard Kalman filter for convenience. Similar accuracy gains are obtained when the error model is integrated into the Kalman filter and applied to the location measurements of vehiclewhose location information is used for validation. We can also observe that there is no observable gain in accuracy when using a higher Gauss–Markov order, sinceprovides almost identical results to. The average measurement error for the standard Kalman filter applied to the location measurements of vehicleis 56 m and 29 m for x and y, respectively, compared to 0.25 m produced by the Kalman filter that is integrated with the autoregression model of order 1. The Kalman filter that was integrated with the autoregression model of order 4 has an average measurement error of 0.24 m, which is very close to that produced by the autoregression model of order 1. This can be further illustrated in Figure 14 and Figure 15, which provide a more detailed view of the behavior of the model for. The fluctuations in error measurements are due to the constant variability in the vehicles’ position measurements, as was found upon further inspection of the original time-series for longitudinal and latitudinal location measurements of both vehicles.

Bottom Line: Vehicle self-localization is an essential requirement for many of the safety applications envisioned for vehicular networks.We prove the existence of correlation between successive measurements in the two datasets, and show that the time correlation between measurements can have a value up to four minutes.Our model can assist in providing better modeling of positioning errors and can be used as a prediction tool to improve the performance of classical localization algorithms such as the Kalman filter.

View Article: PubMed Central - PubMed

Affiliation: College of Information Technology, United Arab Emirates University, Al-Ain 15551, Abu Dhabi. najah@uaeu.ac.ae.

ABSTRACT
Vehicle self-localization is an essential requirement for many of the safety applications envisioned for vehicular networks. The mathematical models used in current vehicular localization schemes focus on modeling the localization error itself, and overlook the potential correlation between successive localization measurement errors. In this paper, we first investigate the existence of correlation between successive positioning measurements, and then incorporate this correlation into the modeling positioning error. We use the Yule Walker equations to determine the degree of correlation between a vehicle's future position and its past positions, and then propose a -order Gauss-Markov model to predict the future position of a vehicle from its past  positions. We investigate the existence of correlation for two datasets representing the mobility traces of two vehicles over a period of time. We prove the existence of correlation between successive measurements in the two datasets, and show that the time correlation between measurements can have a value up to four minutes. Through simulations, we validate the robustness of our model and show that it is possible to use the first-order Gauss-Markov model, which has the least complexity, and still maintain an accurate estimation of a vehicle's future location over time using only its current position. Our model can assist in providing better modeling of positioning errors and can be used as a prediction tool to improve the performance of classical localization algorithms such as the Kalman filter.

No MeSH data available.