A smooth path planning method for mobile robot with A* ant colony optimization was proposed based on dynamic feedback for mobile robot. First of all, in order to overcome the disadvantage about slow convergence speed of ant colony algorithm, simplified A* algorithm was presented to optimize the initial pheromone settings, which was able to solve the blindness of the first search. In this step, the planning path with the minimum value of the valuation function was obtained by the evaluation function of A* algorithm. And the presented multi-evolutionary strategy mechanism which could increase search space was used to strengthen the global search ability of the algorithm. Secondly, in order to further improve the adaptability of algorithm about the problem of local minimum and stagnation in the path planning, the key parameters of the algorithm were systematically analyzed and the closed-loop feedback idea was adopted to adjust the parameters of ant colony optimization algorithm dynamically. Finally, combining with the cubic B spline curve method, the planning path was smoothed to meet the practical movement route of mobile robot. The simulation experiment results showed that compared with traditional ant colony (AC), A* ant colony optimization based on dynamic feedback could reduce 10.4% of the average path cost and shorten 65.8% of the computing time in average. In addition, compared with ant colony system (ACS), the average path cost could be reduced by 5.9%, the calculation time could be shortened by 52.6%. The improved ant colony optimization algorithm could plan a smooth and high quality path in both the dynamic and static environments.