One of the major research fields of the North-Rhine Westphalian research co-operative on artificial intelligence is the development of knowledge based systems which solve complex problems and are tightly coupled with real world processes. Two groups at the universities of Wuppertal and Bonn are co-operating on artificial intelligence applications in high energy physics experiments such as the DELPHI-experiment [1, 2, 3, 8]. The DELPHI-experiment is part of the electron positron accelerator ring LEP at CERN. In the collider ring electrons and their anti-particles, the positrons, are accelerated in opposite directions. Each time an electron collides with a positron, in a so called event, many new and partially short-living particles arise. The finally produced particles can be identified with the help of large detector systems, one of which is the DELPHI detector [5]. The processes after the electron positron annihilation are described by the Standard Model of elementary particle physics. The validation of the Standard Model is a major goal of the LEP experiments.
[1]
K. H. Becks,et al.
B quark tagging using neural networks and comparison with a classical method
,
1992
.
[2]
Charles L. Forgy,et al.
Rete: A Fast Algorithm for the Many Patterns/Many Objects Match Problem
,
1982,
Artif. Intell..
[3]
Wolfram Burgard,et al.
A genetic algorithm for the reconstruction of physical events
,
1991
.
[4]
W. D. Thomas,et al.
The DELPHI detector at LEP.
,
1991
.
[5]
Armin B. Cremers,et al.
DELPHI-EXPERT: An Expert System for Error Diagnosis in High Energy Physics Detectors
,
1992,
IEA/AIE.