In this paper， an unmanned aerial vehicle （UAV） maneuver avoidance decision algorithm based on experiences and intuitions is proposed for threats UAVs encounter during flight processes. Firstly， the situation information obtained during the safety threat avoidance maneuver of the UAV is structured according to the maneuver avoidance area， and the corresponding maneuver decision tag set is recorded. A UAV scenario decision-making knowledge is also established based on the data. Secondly， the long and short-term memory network is used to map the relations between the data scenario and the maneuvering decision in the UAV scenario decision-making knowledge database. Thirdly， according to the principle that human make action decisions based on their intuitions， a UAV maneuver avoidance algorithm is designed based on intuitions. Finally， the simulation analysis is conducted to illustrate the effectiveness of the proposed real-time threat avoidance algorithm for UAVs.