Unmanned Aerial Vehicles (UAVs) are increasingly being adopted for environmental monitoring beyond traditional imaging, yet their use for in-situ air quality assessment remains underdeveloped. This study presents a novel experimental platform specifically designed to evaluate the feasibility and robustness of integrating an infrared (IR) optical sensor on a drone for CO2 detection. A custom laboratory test bench was developed to simulate key drone-mounted environmental conditions, focusing on two dominant disturbances: temperature fluctuations (25–45 ℃) and drone-representative vibrations (10–20 Hz, 0.3–1.3 µm amplitude). The Hamamatsu L15969 light source and InAsSb photovoltaic detector P13243-043CF were selected based on their spectral compatibility with the CO2 absorption band (4.1 µm), compactness, low power consumption, and suitability for UAV payload constraints. Experimental results show that temperature-induced signal distortion is moderate and quantifiable, with an average amplitude decrease of approximately 2% per °C. In contrast, vibrations within the tested range cause severe output incoherence, rendering the sensor signal uninterpretable without effective stabilization. The methodology and results presented here provide actionable guidelines for the integration and deployment of IR spectroscopy sensors on UAVs and establish a foundation for future studies on the combined effects of environmental disturbances. This work advances the applicability of drone-based air quality monitoring by quantifying environmental impacts and proposing practical mitigation strategies.