Search

Three deep tracking methods are presented for the BM@N experiment GEM track detector, which differ in their concepts. The first is a two-stage method with data preprocessing by a directional search in the k-d tree to find all possible candidates for tracks, and then use a deep recurrent neural network to classify them by true and ghost tracks. The second end-to-end method used a deep recurrent neural network to extrapolate the initial tracks, similar to the Kalman filter, which learns necessary parameters from the data. The third method implements our new attempt to adapt the neural graph network approach developed in the HEP.TrkX project at CERN to GEM-specific data. The results of applying these three methods to simulated events are presented. [ABSTRACT FROM AUTHOR]

Tracking particles is a challenging problem in modern high-energy physics detectors producing a vast amount of data such, as experiments on the future NICA collider. Particle track reconstruction is one of the important parts of such experiments, but existing tracking algorithms do not scale well with a growing data stream. In the same time, new effective tracking methods based on graph neural network (GNN) are actively developed and tested in the HEP.TrkX project at CERN. We introduce our GNN approach for the GEM detector of BM@N experiment of the NICA megaproject. This approach is well-adapted for solving the known fake hit problem inherent to strip detectors like GEM with help of minimum branching tree algorithms. Preliminary results for simulated events of the BM@N GEM detector are presented. [ABSTRACT FROM AUTHOR]