Theoretical Approaches For Path Planning Of Intercontinental Formation Flight

The thesis mainly deals with the route optimization for intercontinental formation flight.The problem is fairly complex with the features of aviation related constraints,spherical metric,dynamic network flow and incomplete information,which therefore lead to difficulties in modeling and solving the problem.Four related issues as the problem modeling,the spherical geometric approach of formation path optimiation,the recognition of acceptable formations,and the formation coalition with incomplete information are well studied.First,the definition of equavilent range is introduced starting with the fuel efficiency of formation flight.The problem is formulated into the basic form and the iterative form as well as the constraints within which both aeronautical features and spherical metric are incorporated.The problem is then subdivided into the problem of acceptable formation pattern recoginition and the problem of feasible formation flight scheduling.Therefore the initial intractable problem with huge solution space is divided into a series of finite two-stage problems with rather narrow solution space.Second,the problem is abstracted as the WGSMT problem.The azimuth criteria of the WGSMT Steiner point is also deduced by using which the problem is simplified into a finite geometric planning problem.A construct-revision geometric planning approach is developed aiming at the OAWGSMT route optimization for formation flight.An example is presented to verify the validity of the presented approach.The approximate analytic solution of WGSMT Steiner point is derived on the basis of intercontinental measuring and aerodynamic coupling features.Third,the temporal and spatial boundary of acceptatable formation pattern is derivated.An improved GH-SOM neural network is proposed to recognize acceptable formation patterns.Simulation results indicate that the accuracy and efficiency were far superior than those of GH-SOM and SOM.A multi-agent cooperative formation coalition mechanism is developed to optimize formation schedules based on negotiation.Harsanyi transformation is extended to transform the formation coalition problem under incomplete information into the Bayes-equivalent coalition problem under imperfect information.The sufficient conditions of the existence of a Baysian equilibrium coalition are presented.Simulation results indicate that a better global optimality and fuel efficiency can be achieved compared with those of the proposed-engaged marriage algorithm and simulated annealing algorithm.To be concluded,a comprehensive and systematic research of the route optimization for intercontinental formation flight is done.The achievements may not only build the geometrical basics for the intercontinental formation route optimization and the theoretical foundations for the formation coalition with imcomplete information,but also provide a methodological support for further research on commercial formation flight planning and optimization.