To meet the practical demands of mobile robots in multi-object handling and multi-point inspection tasks, a two-stage fusion algorithm is proposed, combining an improved Ant Colony Optimization (ACO) algorithm with an enhanced Q-Learning approach for efficient multi-objective path planning. In the first stage, the ACO algorithm is enhanced with an optimized distance heuristic function and an elite ant strategy to refine the pheromone update mechanism, thus determining an optimal visiting sequence of targets. In the second stage, an improved Q-Learning algorithm plans the paths between adjacent targets by redefining the state and action spaces, refining the reward function, and accelerating convergence through a combination of the Boltzmann strategy and an adaptive ε-greedy strategy. Furthermore, a path optimization strategy is introduced to enhance the overall path quality. Simulation experiments conducted in three different scenarios demonstrate that the proposed improvements significantly enhance the performance of Q-Learning, and the two-stage fusion algorithm proves to be both effective and feasible for solving multi-objective path planning problems in mobile robotics applications.