The effectiveness of inspection tasks performed by unmanned helicopters during underground inspections is directly influenced by the performance of attitude control. Therefore, it is crucial to optimize the attitude control of unmanned helicopters. Spike neural membrane computing offers Turing computing versatility and high efficiency in various time collaborations. This paper presents a study on attitude optimization control of unmanned helicopters in underground coal mines using spike neural network computing. The approach involves constructing an extended spike neural membrane system based on the attitude dynamics model of an unmanned helicopter, with the aim of further optimizing the system and algorithm. The experimental results demonstrate that the optimized spike neural membrane system is capable of effectively achieving attitude control of the unmanned helicopter under normal conditions. Furthermore, it can successfully maintain the attitude angle within the desired range, even in the presence of wind interference. Additionally, a comparison with traditional sliding mode control and the Designed Optimization Model (OSNMS) reveals that the OSNMS proposed in this paper significantly enhances the attitude control performance of unmanned helicopters.