Focused on the regional boundary guarding task based on a multi-UGVs system, this paper aims to prevent intrusions by external objects into the protected area. By defining the distribution rules for assigning patrol tasks, a hierarchical optimization framework is constructed to optimize the UGV deployment. A genetic algorithm is applied in the outer loop to optimize the distribution of UGV layers. Within the inner loop, a simulated annealing algorithm is applied to optimize the angular distribution of the UGVs. This two-layer framework ensures a tight and effective patrol of the designated area. Simulation results demonstrate that the proposed method effectively reduces the likelihood of external objects intruding into the area.