This study is to formulate an effective robust control methodology tailored for zero-sum differential game scenarios. Initially, the robust zero-sum differential game system model is constructed. A control approach is then devised, with rigorous demonstrations of its asymptotic stability and robustness. The introduction of a data-driven reinforcement learning technique ensures that control outcomes remain impervious to model inaccuracies. Furthermore, an event-triggered mechanism is introduced to augment computational efficiency. Notably, this study marks the first employment of the event-trigger-based adaptive methodology to address challenges posed by the robust zero-sum game issue. Ultimately, efficacy is corroborated through comprehensive simulation analyses.