<p>The directional stability and deviation correction performance of UAV (UAVs) during ground taxiing seriously affect the safety of takeoff and landing of UAVs. In this paper, a dynamic model for ground taxiing and deviation correction of the entire aircraft is established. Under severe crosswind extreme conditions, two different deviation correction parameter control modes, namely, the same gain method and the different gain method, are designed for three different deviation correction subsystems. Based on the radial proxy model, the control parameters are optimized. The results show that the different gain method has the better control effect, and the optimized parameters can greatly improve the deviation correction performance of the subsystem. Considering the impact of sliding speed on the performance of sub deviation correction systems, a joint deviation correction system for UAV ground sliding is designed. Particle swarm optimization algorithm is used to optimize the weight coefficients of the sub deviation correction system, and a joint deviation correction control system that dynamically allocates weights with the UAV sliding speed is obtained. Simulation results show that the joint deviation correction system can achieve better deviation correction control effects using lower sub deviation correction system correction capabilities.</p>