Research On Information Strongly Coupled Mission Planning Methods For Multi-UAV

In recent years,unmanned aerial vehicle(UAV)swarm has been extensively researched,and the UAV swarm technology has become a critical development direction.High autonomous and intelligent combat will become a necessary ability of the future multi-UAV systems.At present,the theory of UAV's autonomy has gradually become mature,and the mission planning technology is the key to achieve the autonomy of UAV swarm.Considering emergency and complex situations in the future battlefield,UAVs should have the ability to carry out mission planning with information couplings,that is,to comprehensively consider the information of the battlefield and the enemy,and to develop effective task execution scheme for each individual in the UAV swarm.In this paper,information transferring coupling,information invoking coupling,and information coupling between UAVs are considered,and information strongly coupled mission planning method is established to improve the mission execution performance with limited computation resources.Firstly,the mathematical model of multi-UAV information coupled mission planning is established.A simplified physical model of UAV and mission scenario,and the basic assumptions and the problem models of each subproblem of mission planning are presented.The model and basic framework of information coupled mission planning are proposed,and the information coupling structure and coupling type classification of multi-UAV mission planning are proposed.Then,a construction method of UAV trajectory under trajectory feature state constraint conditions is established.The theory of trajectory representation is given,and the basic concepts and strict definitions of trajectory feature state variables,trajectory homotopy representation,and trajectory forward kinematics representation are presented.Then the curvature-bounded constraint of trajectory is established,and the generalized circlestraight(CS)curvature-bounded trajectory representation method is given.The theory of trajectory homotopy with bounded curvature,inclusive homotopy of trajectory,and basic curvature-bounded trajectory homotopy are established,and then the method of trajectory classification is proposed.Based on the theory of trajectory envelope,the construction method of path homotopy with bounded curvature is established,including the construction framework of trajectory envelope and the construction method of path homotopy inthe trajectory envelope,which can realize the construction of UAV's trajectory under the constraint conditions.A mission planning method with couplings between UAVs is established based on trajectory planning with expected trajectory feature states.The basic framework of trajectory planning with constant feature state is given,and the coverage of trajectory feature states by the trajectory homotopy is analyzed.The homotopic trajectory planning with constraint of expected trajectory length is analyzed.Basic frameworks of homotopic trajectory planning with expected trajectory length and trajectory planning with expected trajectory length on manifold targets are proposed,and puts forward the corresponding trajectory optimization methods.Aiming at the trajectory planning problem with expected feature states,a trajectory lengthening/shortening method is proposed,and the unrestricted and restricted situations of the problem are analyzed,respectively.The state consensus control model of trajectory feature states is established,a homotopic consensus method is proposed,and a trajectory feature state consensus control algorithm is presented.An information coupled mission planning method based on prediction and verification is proposed.The overall framework of distributed mission planning framework for multi-UAV,as well as the basic frameworks of pre-processing stage,task allocation stage,and post-processing stage,are established.Then,a distributed mission planning method based on prior environment information and iterative verification is established,including environmental description method,basic framework of trajectory prediction,pre-processing method considering environment information prediction,and market mechanism-based allocation algorithm with iterative verification.Among them,the preprocessing method based on prior environmental information combines local trajectory planning with global correction,so as to achieve a compromise between the accuracy of path prediction and calculation amount.By improving the management of UAV onboard computing and communication resource,the iterative verification strategy can re-evaluate the bad planning results and overestimated tasks in the allocation solutions.Finally,an information strongly coupled mission planning method for complex missions is studied.To solve the problem of distributed mission planning of multi-UAV in the scenario including joint tasks,the overall framework of multi-stage trajectory planning,the trajectory smoothing algorithm,and the trajectory correction algorithm for joint tasks are established.By them,the multi-UAV can effectively perform complex tasks withjoint execution requirements.To solve the task allocation problem with predetermined trajectory constraints,a distributed task assignment method based on homotopy method is proposed.The predetermined trajectory generated by the homotopic method is transformed into scheduling solution via the proposed transformation function.Based on this method,the corresponding distributed market mechanism-based task assignment method is established.