Development of Multi Path-Finding Algorithm for Minimum Radiation Exposure in Radiation Controlled Area

Abstract

The International Atomic Energy Agency (IAEA) issues safety guidelines for occupational radiation protection and advocates for reducing worker exposure to the lowest practicable level. Minimizing exposure during movement is crucial for reducing overall occupational exposure. Therefore, a path-finding method is proposed as a solution. Although various studies have proposed different methods to minimize occupational exposure, their applicability is limited in environments with incomplete measurement data. Furthermore, crowded conditions, which often result from the presence of numerous workgroups in radiation areas, increase exposure because of extended walking times. The Conflict-Based Incremental Planner for Minimum Dose (CBIP-MD) was created to address these challenges. This system combines a novel problem definition, data correction methods, time-dependent Lifelong Planning A* (LPA*), a replanner for state space, and a linked list with specific rules. Several tests were conducted to validate this algorithm, including assessments of the radiation weight factor, regularization coefficient, risk analysis, radioactive source estimation, and the state space replanner with the linked list. The findings revealed that adjustment of the radiation weighting factor could reduce occupational exposure by 66.7%, and a low regularization coefficient could further decrease the cumulative dose by accounting for lower radiation levels. The data correction method effectively directed agents to paths with lower exposure doses. In addition to enhancing completeness by reducing computing time by 99%, CBIP-MD maintained occupational exposure doses (0.2%), thereby ensuring optimality. This success was attributed to the application of the state space replanner and the rule-informed linked list.