Aiming at disasters such as fire, earthquake etc. city disaster complex environment, a robot search path algorithm based on ant colony algorithm is put forward in this paper. Suppose environmental was known, we use grid method to divide and model the whole environment space and put forward some improvement in dead circulation, optimizing parameters, stimulating factors for the basic ant colony algorithm, which improved the ant colony algorithm’s robustness effectively. Finally, we obtain search path through numeric experiments in the simulation rescue scenario and make a comparison with original results. Experiment results show the effectiveness of the proposed algorithm and the improved rationality. It will be a good basis for later path planning research in unknown environment.