Aiming at the interception problem of airborne active defense targets, a deep reinforcement learning guidance law based on the double-delay deep deterministic strategy gradient (TD3) algorithm is proposed, which can determine the optimal maneuvering breakout strategy according to the real-time confrontation situation and give the overload command to strike the target accurately while effectively avoiding the threat of defender. The relative motion model is first established under the three-party combat scenario, and then a complete deep reinforcement learning guidance algorithm is constructed by reasonably designing the action space and state space. Lastly, the reward function is designed considering the seeker constraints and energy constraints. The simulation results show that compared with the traditional proportional guidance method, the intelligent game penetration guidance algorithm proposed in this paper has a higher penetration success rate and can accurately hit the target, with good generalization ability and robustness.