This paper proposes a distributed incremental adaptive dynamic programming method to address the optimal cooperative formation control problem for asteroid probes. First, the cost function of spacecraft is formulated by combining the consensus error with the incremental model. Next, the optimal-control theory is used to derive a distributed optimal control law, policy-evaluation and policy-improvement strategies are designed for the distributed control strategy. Finally, by locally and iteratively updating each probe’s neural-network weights, a distributed cooperative formation control law for asteroid probes is obtained. Simulation results verify the effectiveness of the proposed method.