Reentry Flight Capability Assessment And Intelligent Guidance For Aerospace Vehicle

Aerospace vehicle is a reusable concept vehicle that can take off and land horizontally.It has the ability to maneuver in near space and can quickly fly back and forth between ground and space.Compared with the traditional means of delivery,it has higher efficiency and much less expensive.Aerospace vehicle is a combination of hypersonic technology and reusable technology.Relying on the aircraft,a mature and stable transportation system between space and earth can be established,which is of great economic and military value.At present,the frontier research of aerospace vehicle has entered the preliminary test stage.The main design difficulties of its flight profile focus on the reentry process,which is not only faced with dynamic pressure,heat flow,overload and many process constraints,but also has extremely high terminal accuracy requirements.As the concept of artificial intelligence continues to heat up,the design goal of aerospace vehicle also increases the expectation of autonomy and intelligence,which puts forward higher requirements for the ability of real-time information processing during the flight phase.It is quite urgent to put forward more efficient online algorithm.Starting from the reentry process of aerospace vehicle,this paper focuses on the flight capability evaluation and guidance and tries to improve the performance of the relevant algorithms with precise and intelligent goals,so as to provide support for the development of the autonomous decision-making ability of the aircraft.The main research contents include the following aspects:1)For reentry flight capability assessment: visually represent the flight ability of aerospace vehicle through the concept of "reachable domain".Based on commonly used solving method,a new logic is proposed take the no-fly zone impact into consideration.The influence of no-fly zone bring to the reachable domain is discussed by classification.The simulation result is respectively given in every category and shows good robustness.2)To solve the reentry guidance problem,an improved predictor-corrector guidance algorithm is presented by introducing a piecewise objective function.In the early stage the function is defined as range to go while in the terminal stage as drop point error.The modified objective function can guarantee the computational efficiency as well as the guidance accuracy.Furthermore,the no-fly zone avoidance is taken into consideration.In addition to classifying the no-fly zone into two types,an amplitude correction mechanism is employed to cope with the situation where the no-fly zone cannot be avoided only by the bank angle reverse strategy.The Monte Carlo simulation demonstrates that the proposed algorithm is effective on no-fly zone avoidance and has higher guidance accuracy than single objective function method.3)For intelligent algorithms exploring and online operation efficiency enhancing:the "neural network reachable domain generator" and "neural network voyage predictor" are introduced to imply on the flight capability assessment and guidance algorithm.By this change,the operation efficiency of the proposed algorithm is enormously improved,which provides a higher value of online application for further research.