ATLAS

The ATLAS experiment is one of four large experiments currently being conducted at the Large Hadron Collider (LHC) at CERN. After reaching its targeted energy of 13.6 TeV, the next major development is going to increase the luminosity of the LHC. In this High Luminosity (HL) LHC will be about five times the number of simultaneous particle collisions than now and the experiments have to be upgraded to cope with that. ATLAS is a general purpose detector that can observe both, proton-proton and heavy ion collisions. It is build like an onion, different detector types are arranged hermetically in shells around the interaction region.

Development of the LHC over time.
Source

Track Reconstruction

Whenever a particle traverses through matter, it loses some of its energy, which can be measured. Highly energetic particles traverse many detector layers before they either decay, get absorbed, or leave the detector. This string of signals, aligned like pearls on a chain, along the particle trajectory is called a track and it can be used to deduce the momentum, charge, and point of origin of the particle. Unfortunately, it is not a priori known, which hits belong together. The problem of assigning hits to track candidates is called track finding. Track fitting is the process of applying a model to the track candidates to verify them and to calculate the particle properties.


Triggered Readout

After the high luminosity upgrades at the end of this decade, it is expected that there are about 200 proton-proton collisions every 25 ns.

Data produced by the LHC compared with major web services.
Source

Current Activities

Our ATLAS group is working on the implementation of novel track fitting algorithms and modern track finding methods with hardware accelerators (GPUs, FPGAs) for the Event Filter, part of the Trigger and Data Acquisition (TDAQ) system of the ATLAS experiment at the High-Luminosity Large Hadron Collider (HL-LHC). We are working on

• Track finding with Graph Neural Networks on FPGAs
• Novel track fitting algorithms and their acceleration on GPUs (and FPGAs)

You are interested in joining our group to work on one of the largest particle physics experiments that has ever been conducted? Then get in touch!

Past Projects

• Background estimation for searches for the decay Di-Higgs to 4b using Neural Networks
• Measurement of the inclusive triple-differential dijet cross section to disentangle quarks and gluons at the LHC
• Development of a new track trigger concept using three pixel detector layers, aka the Triplet Track Trigger.
• the Hardware Tracking for the Trigger (HTT) project for the ATLAS experiment at the High Luminosity Large Hadron Collider
• the Fast TracKer (FTK) upgrade of the ATLAS detector
• Boosted top quark reconstruction
• Searches for new Physics and Standard Model measurements involving top quarks