Dynamic Sensor Management for Multisensor Multitarget Tracking

Li, Y.; Krakow, Lucas W.; Chong, Edwin; Groom, K.

doi:10.1109/ciss.2006.286683

Cited by 14 publications

(15 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is informative to distinguish between myopic and nonmyopic (also known as dynamic or multistage) resource management, a topic of much current interest (see, e.g., Kreucher et al 2004;Chong 2004, 2006;Bertsekas 2005;Krakow et al 2006;Li et al 2006Li et al , 2007Ji et al 2007). In myopic resource management, the objective is to optimize performance on a perdecision basis.…”

Section: Nonmyopic Adaptive Sensingmentioning

confidence: 99%

Partially Observable Markov Decision Process Approximations for Adaptive Sensing

Chong

Kreucher

Hero

2009

Discrete Event Dyn Syst

View full text Add to dashboard Cite

Adaptive sensing involves actively managing sensor resources to achieve a sensing task, such as object detection, classification, and tracking, and represents a promising direction for new applications of discrete event system methods. We describe an approach to adaptive sensing based on approximately solving a partially observable Markov decision process (POMDP) formulation of the problem. Such approximations are necessary because of the very large state space involved in practical adaptive sensing problems, precluding exact computation of optimal solutions. We review the theory of POMDPs and show how the theory applies to adaptive sensing problems. We then describe a variety of approximation methods, with examples to illustrate their application in adaptive sensing. The examples also demonstrate the gains that are possible from nonmyopic methods relative to myopic methods, and highlight some insights into the dependence of such gains on the sensing resources and environment.

show abstract

Section: Nonmyopic Adaptive Sensingmentioning

confidence: 99%

Partially Observable Markov Decision Process Approximations for Adaptive Sensing

Chong

Kreucher

Hero

2009

Discrete Event Dyn Syst

View full text Add to dashboard Cite

show abstract

“…If object state observations from heterogeneous sensors are available and the environmental conditions or the demand for object state observations change drastically over time, the activation of the most appropriate sensor set can lead to improved results (Suranthiran, Jayasuriya, 2004;van Norden et al, 2005) or the reduction of sensor usage costs (Li et al, 2006).…”

Section: Sensor Schedulingmentioning

confidence: 99%

A time-triggered object tracking subsystem for advanced driver assistance systems

Elmenreich

Koplin

2011

Elektrotech. Inftech.

View full text Add to dashboard Cite

Multi-sensor object tracking is an important feature for advanced driver assistance systems in future automobiles. Most state-of-the-art systems cannot guarantee deterministic processing of the sensor values due to unsynchronized sensing and processing units. To overcome this shortcoming we propose a paradigm shift towards a time-triggered system architecture providing a deterministic bus system, synchronized nodes, and a global time-base. The paradigm shift is supported by results of a simulation of different synchronization and scheduling approaches which suggest that although non-time-triggered approaches perform well in scenarios with low process noise, the time-triggered model becomes advantageous in potentially dangerous scenarios with high dynamics. In order to validate the results of the simulation for real life scenarios, we analyzed test drives derived from a testbed featuring a Volkswagen Touran being equipped with a laser scanner, a stereo camera system, a FlexRay communication system, an object tracking subsystem and a differential GPS system as reference. Zeitgesteuerte Objekterkennung in Fahrerassistenzsystemen.Im Automobil der Zukunft spielen Fahrerassistenzsysteme eine wichtige Rolle. Ein wichtiges Untersystem sind dabei Objektverfolgungssysteme, welche andere Fahrzeuge mit mehreren Sensoren erfassen und deren Position berechnen. Die Architektur der derzeitigen Systeme kann jedoch oft weder Echtzeiteigenschaften noch Determinismus oder synchronisierte Verarbeitung garantieren. Um dieses Problem zu l€ osen, schlagen die Autoren einen Pradigmenwechsel zu einer zeitgesteuerten Architektur vor. Ein simulationsgest€ utzter Vergleich verschiedener Ans€ atze legte die Vermutung nahe, dass die eventgesteuerten Modelle in Szenarien mit niedriger Dynamik bessere Ergebnisse liefern, in potentiell gef€ ahrlichen Szenarien mit hoher Dynamik aber das zeitgesteuerte Modell von Vorteil ist. Um die Realit€ atsn€ ahe der Simulationsergebnisse zu € uberpr€ ufen, wurden beide Ans€ atze in einer Testumgebung mit einem Volkswagen Touran evaluiert. Das Testfahrzeug war hierf€ ur mit einem Laser-Scanner, einem Stereo-Kamera-System, einem FlexRay-Kommunikationssystem, einem Objektverfolgungssystem und einem Differential-GPS-System als Referenz ausgestattet.

show abstract

“…As pointed out in Section 5, exact optimal solutions are practically infeasible to compute. Therefore, recent effort has focused on obtaining approximate solutions, and a number of methods have been developed (e.g., sec [9, 13,14,18,22,23]). Our research contributes to the further development of this thrust by introducing a new approximation method, called nominal belief-state optimization, and applying it to the UAV guidance problem.…”

Section: Introductionmentioning

confidence: 99%

UAV Communication Management and Coordination for Multitarget Tracking

Miller¹,

Harris²,

Chong³

2009

View full text Add to dashboard Cite

The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing aaia SOUIUHS. y^m^niy a,,u maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection ot information, including suggestions for reducing the burden, to the Department of Defense, Executive Service Directorate (0704-0188) Respondents should be aware that notwithstanding any other provision of law. no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR SPONSOR/MONITOR'S ACRONYM(S) AFOSR SPONSOR/MONITOR'S REPORT NUMBER(S) DISTRIBUTION/AVAILABILITY STATEMENTApproved for public release: distribution is unlimited. SUPPLEMENTARY NOTES ABSTRACTThe research conducted under this grant concerned the application of the theory of partially observable Markov decision processes (POMDPs) to the design of guidance algorithms for controlling the motion of unmanned aerial vehicles (UAVs) with on-board sensors to improve tracking of multiple ground targets. While POMDP problems are intractable to solve exactly, principled approximation methods can be devised based on the theory that characterizes optimal solutions. A new approximation method called nominal belief-state optimization (NBO) was proposed. When combined with other application-specific approximations and techniques within the POMDP framework, NBO produced a practical design that coordinated the UAVs to achieve good long-term mean-squared-error tracking performance in the presence of occlusions and dynamic constraints The flexibility of the design was demonstrated by extending the objective to reduce the probability of a track swap in ambiguous situations, with the positive side-effect of improving the mean-squared-error tracking performance as well.

show abstract

Dynamic Sensor Management for Multisensor Multitarget Tracking

Cited by 14 publications

References 22 publications

Partially Observable Markov Decision Process Approximations for Adaptive Sensing

Partially Observable Markov Decision Process Approximations for Adaptive Sensing

A time-triggered object tracking subsystem for advanced driver assistance systems

UAV Communication Management and Coordination for Multitarget Tracking

Contact Info

Product

Resources

About