Conditional Posterior Cramér–Rao Lower Bounds for Nonlinear Sequential Bayesian Estimation

Zuo, Long; Niu, Ruixin; Varshney, Pramod K.

doi:10.1109/tsp.2010.2080268

Cited by 108 publications

(63 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The difference between | − −1 is the positive semi definite matrix. The recursive formula for FIM was given by [44,23]. For the nonlinear filtering problem, considering linear dynamic model, additive Gaussian noise and measurement process the recursive formula reduces to [22],…”

Section: Posterior Cramer Rao Lower Bound (Pcrlb)mentioning

confidence: 99%

Multiple Model Adaptive Nonlinear Filter for Range Estimation Using Bearing Only Tracking

Babulogiah¹,

Jayaraman²,

Sargunaraj³

2017

IJIES

View full text Add to dashboard Cite

This paper discusses about the range estimation using multiple model method for single sensor bearing only tracking (BOT) in 3D. For BOT, the ownship is assumed to take a manoeuvre to gain observability in range and target state. The unknown target range was divided into uniform sub-intervals and nonlinear filters like Extended and unscented Kalman filter (EKF and UKF) with Cartesian and modified spherical coordinates (MSC) were implemented. Comparative results indicate that, UKF in MSC performs better with high computational time. This paper introduces Adaptive nonlinear filter (ANF) to get better range estimate with reduced computational time without affecting the filter performance. It is accomplished by conjoining Cartesian EKF with UKF in MSC, adaptively during the stationary and manoeuvring ownship conditions. The performance comparison was analysed using root mean square error, bias error and computational time. Simulation results reveal that ANF shows better results with reduced computational time.

show abstract

Section: Posterior Cramer Rao Lower Bound (Pcrlb)mentioning

confidence: 99%

Multiple Model Adaptive Nonlinear Filter for Range Estimation Using Bearing Only Tracking

Babulogiah¹,

Jayaraman²,

Sargunaraj³

2017

IJIES

View full text Add to dashboard Cite

show abstract

“…A neural network classifier achieves performance ranging from 77% to 95% on the test data, depending on the type of construction of the location in the building being monitored. [10]. We have proposed and developed a new measure of online tracking performance: the Conditional Posterior Cramér-Rao lower bound (CPCRLB).…”

Section: Feature Identification and Neural Network Classification Formentioning

confidence: 99%

Multimodal Signal Processing for Personnel Detection and Activity Classification for Indoor Surveillance

Varshney¹

2013

Self Cite

View full text Add to dashboard Cite

This goal of this project was to develop novel schemes for the fusion of heterogeneous information. The target application was the detection and classification of personnel activity in both indoor and outdoor environments under dependent observations. We have identified features and designed a classifier that achieves up to 95% classification accuracy on classifying the occupancy with indoor footstep data. MDL-based copula selection strategies are investigated and a detector based on vines is designed that extends previous bivariate copula-based detectors to a multi-sensor application. Our copula-based detectors yield more than 40% improvement over the THIS PAGEThe public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. ABSTRACT Multimodal Signal Processing for Personnel Detection and Activity Classification for Indoor Surveillance Report TitleThis goal of this project was to develop novel schemes for the fusion of heterogeneous information. The target application was the detection and classification of personnel activity in both indoor and outdoor environments under dependent observations. We have identified features and designed a classifier that achieves up to 95% classification accuracy on classifying the occupancy with indoor footstep data. MDL-based copula selection strategies are investigated and a detector based on vines is designed that extends previous bivariate copula-based detectors to a multi-sensor application. Our copula-based detectors yield more than 40% improvement over the conventional data fusion methodologies. We extend our solution to quantized sensor information and demonstrate that injecting controlled noise can dramatically reduce computational complexity with insignificant performance loss. We propose and derive the Conditional Posterior Cramér-Rao Lower Bound (CPCRLB) for online tracking. We demonstrate that the PCRLB-based iterative approach converges quickly with significantly reduced computational cost as compared to a one-shot approach. Detector design in the presence of security threats, such as data falsification attacks to sensor networks, are also addressed. Error-control codes and decoding algorithms are used to reliably classify data in a network containing human agents.

show abstract

“…For example, in [9] an online approximate estimation of the PCRLB for discrete time nonlinear filtering, using the mean and covariance of the estimated online state instead of the true state, is presented. A different approximate recursive formula to calculate conditional PCRLB for nonlinear/non-Guassian Bayesian estimation is presented in [20]. Considering the specific subject of our paper, the inspiring work [4] presents a study, based on CRLB, on AUV positioning uncertainty prediction using LBL and DVL.…”

Section: Related Workmentioning

confidence: 99%

Opportunistic localization of underwater robots using drifters and boats

Arrichiello

Heidarsson

Sukhatme

2012

2012 IEEE International Conference on Robotics and Automation

View full text Add to dashboard Cite

Abstract-The paper characterizes the performance of Autonomous Underwater Vehicle (AUV) localization when the AUV moves in environments where floating drifters or surface vessels are present and can be used for relative localization. In particular, we study how localization performance is affected by parameters e.g. surface object density, their visibility range and their motion. We present a discrete-time nonlinear model for an AUV equipped with onboard sensors and relative positioning information from surface objects (e.g., ranging and bearing) and derive the associated relative Posterior Cramér-Rao Lower Bound. We introduce a probabilistic motion model for the AUV, based on a random direction mobility model, to analyze the expected performance of the localization algorithm in terms of hitting time between the AUV and the surface objects. Finally, an extensive simulation analysis is performed using a discrete time Extended Kalman Filter with Maximum-Likelihood Data Association. As a proof of concept, an AUV equipped with an upward looking sonar is shown to detect a surface vessel and improve its localization estimate.

show abstract

Conditional Posterior Cramér–Rao Lower Bounds for Nonlinear Sequential Bayesian Estimation

Cited by 108 publications

References 15 publications

Multiple Model Adaptive Nonlinear Filter for Range Estimation Using Bearing Only Tracking

Multiple Model Adaptive Nonlinear Filter for Range Estimation Using Bearing Only Tracking

Multimodal Signal Processing for Personnel Detection and Activity Classification for Indoor Surveillance

Opportunistic localization of underwater robots using drifters and boats

Contact Info

Product

Resources

About