When performing unknown target collection tasks, mobile robots frequently face challenges such as unknown environments, lack of information about targets. These challenges can lead robots to overlook corner regions and expand coverage, resulting in low task completion efficiency and long redundant paths. To address these issues, a simultaneous exploration and coverage path planning (SECPP) algorithm is proposed. First, an information gain function consisting of the surrounding environment information and the movement cost is designed. Candidate exploration points are generated by sampling the frontier points, and the point with the maximum information gain is considered as the actual exploration point. Then, a balancing framework is established to evaluate the surrounding environment based on status changes in the environment. If the local environment remains unexplored, the robot continues the exploration task based on the selected exploration point. If the local environment is explored, the algorithm extracts task region information and designs a coverage reward function consisting of the exploration path and guide point to generate a coverage path. The robot follows the path and collects all targets in the region. Finally, the proposed SECPP algorithm is compared with other advanced similar algorithms. The results demonstrate that the SECPP can accomplish the unknown target collection task with a shorter repeated path length, fewer turns, and less time.