Aiming at the problem of multi - robot multi - objective path planning， a multi - robot multi - objective path planning algorithm based on algorithm fusion is proposed. The algorithm adopts the improved A^* algorithm for global path planning， the improved simulated annealing algorithm for planning multiple robots to ensure the shortest path for the farthest robot in the case of combining the optimal global paths， and uses the DWA algorithm for local path planning to realize the local motion obstacle avoidance of robots. In order to verify the feasibility and superiority of the proposed algorithm， simulation experiments are carried out. The experimental results show that the improved A^* algorithm reduces the number of turns， turning angles， traversal nodes， and planned distance by 16.66%， 41.24%， 23.3%， and 0.77% respectively compared to the classical A^* algorithm. The improved simulated annealing algorithm reduces simulation time， longest path length， and total path by 12.09%， 22.26%， and 5.74% respectively compared to the traditional simulated annealing algorithm. The DWA algorithm can achieve multi robot local path planning and obstacle avoidance.