Research On Passive Homing Missiles Guidance Laws With Observability Enhancement

Passive homing missiles have the advantages of great survivability and high cost-effectiveness,since they can generate the guidance commands by sensing the target reflection or radiation energy.To satisfy the requirements of modern warfare on guidance precision and strike performance,advanced guidance laws considering impact angle and/or impact time are widely studied.Because advanced guidance laws require information about target motion states and passive sensors can only directly measure bearing information,it is needed to conduct target motion analysis to estimate target motion states using bearings-only measurements,thus the observation performance on targets directly affects the guidance precision.Considering the observability of targets,this thesis conducts studies on guidance law under multiple constraints,multi-missile cooperative guidance law,and multiple targets attack assignment for acquiring higher estimation accuracy of the target motion states to enhance the guidance precision and strike performance of a single missile or multi-missile swarms.According to the guidance process of the passive homing missile,the basic models including missile-target relative kinematics,impact angle constraint and field-of-view constraints are constructed.For target motion analysis,typical target motion models and state equations and measurement equation in tracking coordinate system are presented,and the extended Kalman filter is employed to generate nonlinear estimations of target motion states.Further,covariance intersection algorithm is introduced to achieve data fusion of different missiles observation information in multi-missile cooperative attack task.To achieve guidance under impact angle,field-of-view and loading factor constraints,a modified circular navigation guidance law(MCNGL)without using time-to-go estimation is proposed.Firstly,the desired biarc flight trajectory is designed considering the initial path angle and terminal impact angle,and the feasible solution set of biarc trajectories under trajectory coherence,loading factor and field-of-view is derived.Then,the gain variable proportional guidance law and sliding-mode guidance law are tailored according to the different characteristics of two arcs,and they constitute the MCNGL.Numerical simulation results demonstrate that MCNGL can satisfy the complex guidance constraints,and it can provide comparable results to optimal guidance law in terms of optimality.To satisfy the impact angle constraint and improve estimation accuracy,a two-phase guidance framework consisting of optimal observation phase and constrained attack phase is proposed.In the optimal observation phase,the non-integral observability index based on determinant of Fisher information matrix is constructed and it is maximized to derive the optimal leading angle considering the field-of-view.Then,a sliding-mode guidance law is designed to track the desired leading angle based on the finite-time convergence sliding-mode control.In the followed attacking phase,the MCNGL is applied to satisfy the given guidance constraints.Simulation results demonstrate that the proposed guidance law can reduce the estimation errors of target states and achieve better terminal performances including miss distance and impact angle accuracy.For multi-missile cooperative attack tasks,a leader-follower cooperative guidance frame considering observability of target motion states is proposed,and a cooperative guidance law considering multiple-missile observation is presented.The leader employs the two-phase guidance law,and the follower employs the sliding-mode guidance law based on the coordinated variable controlled by leading angle.The cooperative observation configuration is controlled by setting the relative errors of missile-target bearing as coordinated variables and the cooperative attack time is controlled by setting the relative errors of missile-target distance as coordinated variables.Comparison studies are provided to show that the proposed method can satisfy the cooperative attack requirements and can improve the target state estimation accuracy and cooperative strike accuracy.For the multi-target cooperative attack mission,the effects of cooperative observation on the attack effectiveness are considered,and the target assignment is formulated as a nonlinear integer optimization problem.An effective target assignment method using combinatorial bids-based auction is proposed.The reward allocation mechanism of the cooperative missiles is constructed to support combinatorial bids,and the auction iteration strategy and bid updating strategy are presented to acquire the equilibrium results.Numerical simulations show the proposed method has comparable optimality and better computation efficiency than genetic algorithm.And the comparison studies demonstrate that cooperative observation can improve the cooperative attack effectiveness of the target assignment results.