The PANDA experiment at FAIR offers unique possibilities for performing hyperon physics. The detector will enable the reconstruction of both hyperon and antihyperon, which will be created together in proton-antiproton collisions. This enables investigations of the strong interaction in the non-perturbative regime. Due to their relatively long-lived nature, the hyperons impose a particular challenge on the track reconstruction and event building. In order to exploit the large expected reaction rates to the fullest, PANDA will utilize a fully software-based event filtering. Therefore, reconstructing hyperons for such a filter requires online track reconstruction that can handle particles created a measurable distance away from the interaction point and, at the same time, operate on free streaming data is needed. Until antiprotons are available at PANDA, a part of the hyperon program can be carried out with the predecessor, PANDA@HADES using a proton beam. In this thesis, investigations of the detector signatures from the decay channels Λ → pπ-, Ξ- → Λπ- and Ω- → Λ K- produced in YbarY reactions are presented. The detector signatures guide the subsequent track reconstruction algorithms. A candidate for online track reconstruction algorithms on free streaming data based on a 4D Cellular Automaton has been developed and is benchmarked. It utilizes information from the PANDA straw tube tracker and is agnostic to the point of origin of the particle. The track reconstruction quality assurance procedure and results from the tracking at different event rates are also presented. Finally, extrapolation algorithms for including hit information from additional detectors in the tracks are outlined. In order to maximize the potential of the predecessor experiment PANDA@HADES, a kinematic fitting procedure has been developed for HADES that combines geometric the decay vertex information of neutral particles and track parameters such as momentum. Benchmark studies on simulated data from the channel p(3.5 GeV)p → ΛK+p are presented as well as tests of the kinematic fit on experimental data from 2007.

Next-generation DIRC detectors, like the PANDA Barrel DIRC, with improved optical designs and better spatial and timing resolution, require correspondingly advanced reconstruction and PID methods. The investigation of the PID performance of two DIRC counters and the evaluation of the reconstruction and PID algorithms form the core of this thesis. Several reconstruction and PID approaches were developed, optimized, and tested using hadronic beam particles, experimental physics events, and Geant simulations. The near-final design of the PANDA Barrel DIRC was evaluated with a prototype in the T9 beamline at CERN in 2018. The analysis finds excellent agreement between the experimental data and the Geant simulations for all reconstruction algorithms. The best PID performance of up to $5.2 \pm 0.2$ s.d. $\pi$/K separation at 3.5 GeV/c, was obtained with a time imaging PID method. The PANDA Barrel DIRC simulation, as well as the reconstruction and PID algorithms, were evaluated using experimental data from the GlueX DIRC as part of the FAIR Phase-0 program. The performance validation was carried out using physics events of the GlueX experiment and simulations. The initial analysis results of the commissioning dataset show a $\pi$/K separation power of up to 3 s.d. at a momentum of 3.0-3.5 GeV/c, obtained using a geometric reconstruction algorithm.