 |
Gunter van Keuk
|
 |
| data fusion subpanel architecture of the Joint Directors of Laboratories |
When I read this article in AES Magazine, I was interested, as I had been working on various applications of multiple sensor data fusion at Hughes Aircraft (later Raytheon) for the US DoD during the same time period in air defense, radar, EW, and sonar. I wrote several articles for the data fusion subpanel of the Joint Directors of Laboratories in the DoD regime in the 1980-1990s, using Neural Networks, Baysian, Kalman and other methodologies with a TADIL J network. While I followed Blackman and Linas and Hall, I wish I had been more aware of van Keuk's work, which mostly arrived later. Back then most sensors were predefined, and feedback from fusion processing was limited. The big breakthrough in sensor data fusion came at the JHU/APL with Peri's Joint
Composite Combat ID Analysis Testbed (CAT).
This article provides a retrospective on the life and work of Günter van Keuk, a pioneer in sensor data fusion and tracking research in Germany. Some key points:
- Van Keuk was a physicist who
joined the Research Institute for Radio and Mathematics (FFM) in 1968.
He founded the department for Sensor Data Processing and Control Methods
in 1975.
- At FFM, van Keuk developed
foundational techniques for fusing data from multiple radar sensors to
track targets. This included some of the earliest work on electronically
steerable phased array radars.
- He promoted adaptive tracking
methods that optimized radar resources based on feedback from the
tracking system. His research enabled radar networking and data fusion
capabilities.
- Van Keuk published influential
papers on phased array tracking, multi-hypothesis tracking, sequential
track extraction, and other topics. He collaborated with international
experts like Sam Blackman.
- After the Cold War ended in
1990, van Keuk became more involved in international conferences and
societies. This allowed his innovations in sensor fusion to gain wider
recognition.
- The article chronicles van
Keuk's career from the Cold War era to his retirement in 2001, assessing
his pioneering contributions during a period of rapid technological
change. It emphasizes his lasting impact on data fusion research in
Germany and beyond.
Abstract: Multiple sensor data fusion and adaptive sensor resources management are key technologies for realizing modern aerospace and electronic systems, researched by the information fusion community. In October 2023, on the 20th anniversary of the death of Günter van Keuk, one of its pioneers, we recall aspects of the early history of this branch of applied computer science in the German defense domain.
In 2023 also the Fraunhofer Institute for Communications, Information Processing, and Ergonomics (FKIE) looks back on 60 years of one of its roots, the Research Institute for Radio and Mathematics (FFM), van Keuk's scientific home, as well as on 60 years of target tracking in Germany, which has led to FFM's foundation in 1963.
URL: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10258285&isnumber=10311344
W. Koch, "Günter van Keuk (1939–2003) and the Evolution of Data Fusion
Research in Germany," in IEEE Aerospace and Electronic Systems Magazine,
vol. 38, no. 11, pp. 4-15, Nov. 2023, doi: 10.1109/MAES.2023.3317880.
Background
Prof. Dr. Wolfgang Koch is Head of Department of
sensor data fusion at FFM. He coauthored papers with van Keuk. His work at the
Fraunhofer Institute for Communication, Information Processing and Ergonomics FKIE helps to detect and avoid threats. We obtain, transmit,
process and protect information and present it in an understandable way.
Our systems are used by the Bundeswehr, civil security institutions and
industry.
Seamless safety assistance systems
The
defense against threats - for example from terrorism - depends on
intelligent situation reports. Advanced tracking and data fusion methods
evaluate data from networks of different sensors and thus increase
situational awareness. Robot systems continue to provide data even when
missions are too dangerous for humans.
Intelligent decision support systems
Intelligent
decision support systems make relevant information quickly accessible
and consistently distribute merged information to the right places. In
addition, we develop human-machine interfaces that enable people to work
with the systems intuitively and effectively.
Robust and secure communication systems
For
modern crisis operations, we develop robust networks that work reliably
even when radio propagation is problematic. Reliable methods classify
transmitters even under difficult conditions. The security of critical
network infrastructures is the focus of cyber defense research.
Publications authored by Van Keuk:
W. Koch and G. Van Keuk, "Multiple hypothesis track maintenance with possibly unresolved measurements," in IEEE Transactions on Aerospace and Electronic Systems, vol. 33, no. 3, pp. 883-892, July 1997, doi: 10.1109/7.599263.
Abstract: In surveillance problems dense clutter/dense target situations call for refined data association and tracking techniques. In addition, closely spaced targets may exist which are not resolved. This phenomenon has to be considered explicitly in the tracking algorithm. We concentrate on two targets which temporarily move in close formation and derive a generalization of MHT methods on the basis of a simple resolution model.
URL: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=599263&isnumber=13075
G. Van Keuk, "Sequential track extraction," in IEEE Transactions on Aerospace and Electronic Systems, vol. 34, no. 4, pp. 1135-1148, Oct. 1998, doi: 10.1109/7.722699.
Abstract:
Sensors like radar or sonar usually produce data on the basis of a
single frame of observation: target detections. The detection
performance is described by quantities like detection probability Pd and
false alarm density f. A different task of detection is formation of
tracks of targets unknown in number from data of multiple consecutive
frames of observation. This leads to quantities which are of a higher
level of abstraction: extracted tracks. This again is a detection
process. Under benign conditions (high Pd, low f and well separated
targets) conventional methods of track initiation are recommended to
solve a simple task. However, under hard conditions the process of track
extraction is known to be difficult. We here concentrate on the case of
well separated targets and derive an optimal combinatorial method which
can be used under hard operating conditions. The method relates to MHT
(multiple hypothesis tracking), uses a sequential likelihood ratio test
and derives benefit from processing signal strength information. The
performance of the track extraction method is described by parameters
such as detection probability and false detection rate on track level,
while Pd and f are input parameters which relate to the signal-to-noise
interference ratio (SNIR), the clutter density, and the threshold set
for target detection. In particular the average test lengths are
analyzed parametrically as they are relevant for a user to estimate the
time delay for track formation under hard conditions.
URL: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=722699&isnumber=15626
What is Sensor Data and Information Fusion? Sensor
Data and Information Fusion is the process of combining incomplete and
imperfect pieces of mutually complementary sensor data or non-sensor
information in such a way that a better understanding of an underlying
real-world phenomenon is achieved. Typically, this insight is either
unobtainable otherwise or a fusion result exceeds what can be produced
from a single sensor output or other information source in accuracy,
reliability, or cost. Appropriate collection, registration and
alignment, stochastic filtering, logical analysis, space-time
integration, exploitation of redundancies, quantitative evaluation, and
appropriate display are part of sensor data fusion as well as the
integration of related context information.
G. van Keuk and S. S. Blackman, "On phased-array radar tracking and parameter control," in IEEE Transactions on Aerospace and Electronic Systems, vol. 29, no. 1, pp. 186-194, Jan. 1993, doi: 10.1109/7.249124.
Abstract: Based on a simple model of a ground-based phased-array radar used for a multiple-target surveillance task, beam scheduling, positioning, and radar parameters like signal-to-noise ratio, track sharpness, and detection threshold have been optimized with respect to the radar/computer load induced by tracking. From this the minimum energy necessary for track maintenance during surveillance can be derived.
URL:
https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=249124&isnumber=6377
G. Van Keuk, "Multihypothesis tracking using incoherent signal-strength information," in IEEE Transactions on Aerospace and Electronic Systems, vol. 32, no. 3, pp. 1164-1170, July 1996, doi: 10.1109/7.532278.
Abstract: Multihypothesis tracking (MHT) methods have been suggested to track targets on the basis of noisy target locations over a background of false detections. Assuming a standard fluctuation model of the target cross section MHT methods are generalized to additionally incorporate signal-strength information (incoherent) provided by the detection process. The detection threshold has been selected to optimize track performance. The impact of the signal-to-noise ratio (SNR) on the tracking task has been analyzed in order to identify SNR conditions which call for refined tracking methods like MHT. By this we also get an estimate of the improvement achieved by MHT techniques over more standard approaches.
URL:
https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=532278&isnumber=11084
G. Van Keuk, "MHT extraction and track maintenance of a target formation," in IEEE Transactions on Aerospace and Electronic Systems, vol. 38, no. 1, pp. 288-295, Jan. 2002, doi: 10.1109/7.993248.
Abstract: Tracking of a group of targets temporarily operating in formation is relevant for military air-surveillance. An optimal sequential multiple hypothesis tracking (MHT) track extraction and track maintenance method for well-separated single targets embedded in clutter has been described in an earlier paper. However, applied to a group of closely spaced targets, its performance is not acceptable due to data miscorrelation. Here the approach has been generalized to include extraction and maintenance of a maneuvering formation in the presence of clutter. The unknown number of targets and imperfect detection when the group newly appears is modeled within the MHT framework. The generalized method solves the problem very well and can also track a formation after the targets separate.URL:
https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=993248&isnumber=21418
W. Koch and G. Van Keuk, "Multiple hypothesis track maintenance with possibly unresolved measurements," in IEEE Transactions on Aerospace and Electronic Systems, vol. 33, no. 3, pp. 883-892, July 1997, doi: 10.1109/7.599263.
Abstract: In surveillance problems dense clutter/dense target situations call for refined data association and tracking techniques. In addition, closely spaced targets may exist which are not resolved. This phenomenon has to be considered explicitly in the tracking algorithm. We concentrate on two targets which temporarily move in close formation and derive a generalization of MHT methods on the basis of a simple resolution model.
G. Van Keuk, "Sequential track extraction," in IEEE Transactions on Aerospace and Electronic Systems, vol. 34, no. 4, pp. 1135-1148, Oct. 1998, doi: 10.1109/7.722699.
Abstract: Sensors like radar or sonar usually produce data on the basis of a single frame of observation: target detections. The detection performance is described by quantities like detection probability Pd and false alarm density f. A different task of detection is formation of tracks of targets unknown in number from data of multiple consecutive frames of observation. This leads to quantities which are of a higher level of abstraction: extracted tracks. This again is a detection process. Under benign conditions (high Pd, low f and well separated targets) conventional methods of track initiation are recommended to solve a simple task. However, under hard conditions the process of track extraction is known to be difficult. We here concentrate on the case of well separated targets and derive an optimal combinatorial method which can be used under hard operating conditions. The method relates to MHT (multiple hypothesis tracking), uses a sequential likelihood ratio test and derives benefit from processing signal strength information. The performance of the track extraction method is described by parameters such as detection probability and false detection rate on track level, while Pd and f are input parameters which relate to the signal-to-noise interference ratio (SNIR), the clutter density, and the threshold set for target detection. In particular the average test lengths are analyzed parametrically as they are relevant for a user to estimate the time delay for track formation under hard conditions.
G. van Keuk, "Multihypothesis tracking with electronically scanned radar," in IEEE Transactions on Aerospace and Electronic Systems, vol. 31, no. 3, pp. 916-927, July 1995, doi: 10.1109/7.395247.
Abstract: Multihypothesis tracking (MHT) methods, well known for conventional radar data-processing, are generalized for phased-array radar application. We consider single targets in a cluttered environment. Suitable revisit and beam positioning algorithms are derived and optimized regarding the energy spent for tracking. In particular the destabilization of tracks due to clutter interference is analyzed. A comparison with a conventional algorithm (probabilistic data association filter (PDAF)) shows the saving of radar load which can be achieved by the proposed method.
R. Baltes and G. van Keuk, "Tracking multiple manoeuvering targets in a network of passive radars," Proceedings International Radar Conference, Alexandria, VA, USA, 1995, pp. 304-309, doi: 10.1109/RADAR.1995.522563.Abstract: This paper addresses the problem of tracking multiple targets in a net of passive (angle only) 2D- of 3D-radar (azimuth or azimuth and elevation). No decentralised local bearing tracks are built, but measurements are fused on a central level. No signal or signal following information is used. We assume multiple coverage, high probability of detection and consider resolution conflicts of data and false alarms. A basic algorithm used to process asynchronous decentralised bearings is given. An algorithm for a full online tracking system with track initiation and deletion is described. Its central part is the global deghosting algorithm. It inherently works as the global (space/time) data for the track assignment algorithm. Simulation results of a dynamic scenario are given.
No comments:
Post a Comment