Abstract: ‍ ‍Aerial intelligent reflecting surface （AIRS） combines the advantages of aerial platforms and intelligent reflecting surface， which can be flexibly deployed in various wireless network topologies to improve system performance. Motivated by the remarkable advantages of AIRS， to address the challenge of poor transmission quality and low-security performance in wireless transmission caused by obstacle blocking and eavesdropping， this study proposes a secure and robust resource allocation algorithm for AIRS-assisted multiple-input single-output communication systems. Considering the impact of the imperfect channel state information of eavesdropping channels， a multivariable robust resource allocation problem was formulated for jointly designing the active beamforming of the base station， passive beamforming of multiple AIRSs， and deployment locations of multiple AIRSs to maximize the worst-case sum secrecy rate of communication systems while satisfying the maximum transmission power constraint of the base station， phase shift constraints of the AIRS， deployment location constraints of the AIRS， and minimum secrecy rate constraints of legitimate users. However， the formulated resource allocation problem was naturally non-convex because of the high degree of coupling and non-linear relationship between these optimization variables， making the aforementioned problem impossible to solve directly. To address the abovementioned sophisticated non-convex problem， first， the block coordinate descent method was used to decompose the original problem into three manageable subproblems， including active beamforming design， passive beamforming design， and deployment location design. Then， these non-convex subproblems were transformed into convex optimization problems by applying variable relaxation， penalty functions， and successive convex approximation methods. Finally， an overall algorithm was proposed to solve the original optimization problem. The algorithm gradually converges to a suboptimal solution of the original problem by solving each subproblem in alternating iterations. Simulation results show that the proposed resource allocation algorithm can effectively improve the worst-case sum secrecy rate of the communication system and has better robustness than other benchmark algorithms. These results validate the significant potential of AIRS in enabling secure wireless communications and emphasize the importance of designing deployment locations and passive beamforming for AIRS.