Optimization Methods for Track Fitting in the Active-Target Time Projection Chamber

Local and global optimization methods were evaluated for fitting spiral tracks in the Active-Target Time Projection Chamber (AT-TPC) at the National Superconducting Cyclotron Laboratory at Michigan State University. The AT-TPC is a gas-filled chamber that acts as both the target and the detector for studying nuclear reactions. Using this, we can reconstruct three-dimensional spatial tracks of the reaction products. Track-fitting methods were tested on data from the $^{46}$Ar(p, p) and $^{40}$Ar(p, p) experiments that ran in 2015. We aim to improve on the current fitting method in the analysis software, the naive Monte Carlo algorithm. Various optimization methods were tested, with notable success using global optimization methods, specifically, differential evolution and basin hopping.  Local optimization methods were tested in the basin hopping algorithm in order to examine the performance impact. Results will be presented that compare the accuracy, robustness to noise, and time efficiency of each method.