Based on the respective strengths of multi-agent and differential evolution algorithm, multi-agent’s ability of
sensing the environment and reacting to the environment and differential evolution’s capacity of the speed and good global
optimization are fully combined, and multi-agent differential evolution algorithm is proposed. The proposed differential
evolution operator is introduced to improve the update speed of agent, and the diversity of the population is kept. The
orthogonal crossover operator is imported to ameliorate cooperation characteristic and compete with their neighbors
effectively. The local optimization operator is applied to improve searching precision. Several classic test functions are
tested, and the results show that the proposed algorithm can improve the global convergence ability.