Abstract: With the development of intelligent mining, autonomous emergency braking system is equipped on the mining transportation vehicles. However, since open-pit mining areas are often accompanied by unstructured roads and extreme weather such as rain, snow and fog, as well as poor road adhesion coefficient under the influence of weather, these have a great impact on the perception and decision-making system of autonomous emergency braking system, which often leads to collision or braking under safe conditions. This paper conducts investigation on the intended functional safety issues of the autonomous emergency braking system for mining transport vehicles during open-pit mining operations, aiming to address the safety issues caused by system failures due to insufficient system functionality triggered by risk scenarios. Firstly, this paper employs the systematic theory process to identify the scenarios that cause unsafe behaviors in the autonomous emergency braking system and proposes improvement strategies for the perception and decision-making systems. Secondly, it adopts an image enhancement method based on dark channel prior and an Extended Kalman Filter fusion algorithm to address the issue of inaccurate perception in adverse weather conditions. A collision time model that considers factors such as slope, coefficient of friction, slip ratio, and vehicle load is proposed to enhance the adaptability of the decision-making system to mining operational environments. Finally, a joint simulation platform is established based on MATLAB/Simulink, TruckSim, and PreScan, and the most hazardous test cases are selected for testing. The results indicate that in non-intersection scenario tests, collisions occurred in systems where only perception was improved, and in?intersection?scenario tests, these systems also exhibited lower safety margins. In contrast, systems where only the decision-making component was improved resulted in inaccurate perception data due to environmental factors such as weather and lighting, leading to overly conservative decisions and premature braking. Systems with comprehensive improvements in both perception and decision-making can effectively avoid collisions without triggering braking too early. This paper demonstrates that comprehensive improvements to both perception and decision-making in automatic emergency braking systems can effectively address the issues of premature and delayed braking, enhancing the system's safety and adaptability.