The detection of terrestrial planets is an important field of deep space exploration,and the trajectory tracking control of the planetary probe entry segment can effectively improve the final landing accuracy of the planetary probe.During the flight of the planetary entry stage,the probe faces a harsh and complex aerodynamic environment,and its flight trajectory is affected by many disturbances such as the uncertainty of atmospheric density,the uncertainty of the dynamic model,and the uncertainty of aerodynamic parameters.Effective guidance and control of the probe is required during the planetary entry segment.In this paper,under the comprehensive consideration of various uncertain disturbances,based on the radial basis function neural network uncertainty estimation method and the sliding mode control method,several composite control methods are constructed to track the trajectory of the planetary probe.issue research.The details are as follows:1.Considering the uncertainty of atmospheric density as the main source of disturbance during the landing of planetary probes,a trajectory tracking control method based on radial basis function neural network and fractional sliding mode is proposed.Based on the three-degree-of-freedom dynamic model of the probe entry segment,the sliding mode control is used to track the nominal trajectory of the planetary probe entry segment,the radial basis function neural network is used to estimate and compensate the uncertainty of atmospheric density,and fractional calculus is introduced into the sliding mode control reaching law,the switching gain of the sliding mode control caused by the uncertainty of the atmospheric density is reduced,and the buffeting of the system is weakened.The method is applied to the simulation of the Mars landing scene.The control method can accurately track the landing trajectory of the probe under the uncertainty of the unknown atmospheric density,so that the planetary probe can reach the parachute opening point with high precision.2.On the basis of the above research,further considering that the interference source is the uncertainty of atmospheric density and aerodynamic parameters,a trajectory tracking control method for the entry segment of the planetary probe is designed.The traditional linear sliding mode is improved,and a non-singular fast terminal sliding mode controller is designed to ensure that the trajectory tracking system of the entry segment has a fast convergence speed and the tracking error converges to zero in a finite time.Combined with the radial basis function neural network,the total uncertainty online real-time estimates.The trajectory tracking control of the entry segment of the planetary probe with strong robustness,high control accuracy and limited time convergence of tracking error is realized.3.On the basis of the above research,the interference source is further considered as the uncertainty of atmospheric density,aerodynamic parameters and dynamic model,and a trajectory tracking control method for the entry segment of the planetary probe is designed.In order to improve the approximation ability of radial basis function neural network,the improved particle swarm algorithm with optimized inertia weight is used to optimize the structural parameters of radial basis function neural network,and then the optimized radial basis function neural network is used to estimate and compensate the total uncertainty.Combined with non-singular fast terminal sliding mode,finite-time trajectory tracking control is realized.The validity and accuracy of the method are verified by the simulation of the Mars entry stage.1000 times of Monte Carlo parachute simulation experiments are carried out under the condition that the initial state is uncertain,which verifies the robustness of the method. |