Condition-based maintenance can arrange preventive maintenance according to the operating condition of the equipment,which is an effective way to solve the degradation failure and ensure the availability of equipments.In this paper,we focuse on the core components of the UAVs and develop the condition-based maintenance model as well as replacement policy with the method of continuous-time Markov chain by dividing the deterioration into different states happened in the operational process.In the model,the maximal steady state availability of the system is used as the indicator to arrange maintenance,while the impact of the such factors as average time of the deterioration between two adjacent stages,the average detection time,the average time of preventive maintenance and the time for part replacing and etc,is also taken into account.Ultimately,the optimal frequency of testing system and the threshold of condition-based maintenance as well as the replacement policy are determined by using a recursive algorithm to solve the model according to the principle of statistical equilibrium in the steady state of Markov process.The result shows that the presented the model can describe the degradation process effectively,and can be used to optimize the maintenance policies.