With increasing complexity of the operational environment, aiming at the target assignment problem of the manned/unmanned cooperative air-to-ground attack mission, the collaborative target assignment model of UCAVs is built up through analyzing enemy threat coefficient and Unmanned Combat Aerial Vehicles’ (UCAVs) damage effectiveness. An improved Grey Wolf Optimization (GWO) algorithm is proposed to improve quality and speed of the solution, which uses triploid grey wolf gene strategy to increase initial population diversity, improves the convergence speed by weighted combination location updating strategy and introduces nonlinear factor as well as different mutation operators to achieve the balance between global search and local search. By simulation experiments, the feasibility and superiority of the algorithm are verified. And compared with GWO algorithm, the improved GWO algorithm’s optimization process is faster and more accurate, which can well adapt to the complex battlefield environment and meet the requirements of target assignment in the dynamic combat process.