In this paper, based on adaptive genetic algorithm and grey relation degree, a new method for solving multi-objective problem is put forward. First, the best solution of every objective in the middle of the multi-objective is solved and they are looked on as referenced vector. Second, the grey relation degree between every individual and the referenced vector is solved and the grey relation degree is acted as fitness of the individual. At last, the pareto optimal sets are solved by means of adaptive genetic algorithm. The variety of population is kept by means of adaptive probability of crossover and mutation. Simulation examples show the effectiveness of the proposed approach.