Recorded recurrent deep reinforcement learning guidance laws for intercepting endoatmospheric maneuvering missiles

This work proposes a recorded recurrent twin delayed deep deterministic(RRTD3)policy gradient algorithm to solve the challenge of constructing guidance laws for intercepting endoatmospheric maneuvering missiles with uncertainties and observation noise.The attack-defense engagement scenario is modeled as a partially observable Markov decision process(POMDP).Given the benefits of recurrent neural networks(RNNs)in processing sequence information,an RNN layer is incorporated into the agent’s policy network to alleviate the bottleneck of traditional deep reinforcement learning methods while dealing with POMDPs.The measurements from the interceptor’s seeker during each guidance cycle are combined into one sequence as the input to the policy network since the detection frequency of an interceptor is usually higher than its guidance frequency.During training,the hidden states of the RNN layer in the policy network are recorded to overcome the partially observable problem that this RNN layer causes inside the agent.The training curves show that the proposed RRTD3 successfully enhances data efficiency,training speed,and training stability.The test results confirm the advantages of the RRTD3-based guidance laws over some conventional guidance laws.

Online task planning method of anti-ship missile based on rolling optimization

Based on the wave attack task planning method in static complex environment and the rolling optimization framework, an online task planning method in dynamic complex environment based on rolling optimization is proposed. In the process of online task planning in dynamic complex environment,online task planning is based on event triggering including target information update event, new target addition event, target failure event, weapon failure event, etc., and the methods include defense area reanalysis, parameter space update, and mission re-planning. Simulation is conducted for different events and the result shows that the index value of the attack scenario after re-planning is better than that before re-planning and according to the probability distribution of statistical simulation method, the index value distribution after re-planning is obviously in the region of high index value, and the index value gap before and after re-planning is related to the degree of posture change.

Time-varying parameters estimation with adaptive neural network EKF for missile-dual control system

In this paper, a filtering method is presented to estimate time-varying parameters of a missile dual control system with tail fins and reaction jets as control variables. In this method, the long-short-term memory(LSTM) neural network is nested into the extended Kalman filter(EKF) to modify the Kalman gain such that the filtering performance is improved in the presence of large model uncertainties. To avoid the unstable network output caused by the abrupt changes of system states,an adaptive correction factor is introduced to correct the network output online. In the process of training the network, a multi-gradient descent learning mode is proposed to better fit the internal state of the system, and a rolling training is used to implement an online prediction logic. Based on the Lyapunov second method, we discuss the stability of the system, the result shows that when the training error of neural network is sufficiently small, the system is asymptotically stable. With its application to the estimation of time-varying parameters of a missile dual control system, the LSTM-EKF shows better filtering performance than the EKF and adaptive EKF(AEKF) when there exist large uncertainties in the system model.

Active Control of Initial Disturbances for Rockets and Missiles

The active control theory and methods of initial disturbances for rockets and missiles are investigated. The rocket or missile/launcher is simplified as a flexible beam excited by a moving varying velocity rigid body which has two points in contact with the beam. The control force is applied at the supporting point on the beam. Active control strategies based on optimal control theory are proposed and computer simulation is carried out. Simulation results are consistent with the theoretical results, and show that the active control strategies proposed can accomplish the purpose to control the initial disturbances actively. The results show that active control of initial disturbances for rockets and missiles is feasible for application.

Mid-Course Guidance Law for Long Range Multi-Purpose Missiles

A useful mid course guidance law for long range multi purpose missiles is proposed. The law consists of two parts. One is used in the horizontal guidance plane, and the other in the vertical guidance plane. The mid course is combined with variable structure proportional navigation in the homing phase. Simulation results show that the proposed guidance law can meet the demands for starting up the homing phase and so the handing over between mid course guidance and terminal guidance can be carried out successfully.

Saturation attack based route planning and threat avoidance algorithm for cruise missiles

According to the characteristic of cruise missiles,navigation point setting is simplified,and the principle of route planning for saturation attack and a concept of reference route are put forward.With the help of the shortest-tangent idea in route-planning and the algorithm of back reasoning from targets,a reference route algorithm is built on the shortest range and threat avoidance.Then a route-flight-time algorithm is built on navigation points.Based on the conditions of multi-direction saturation attack,a route planning algorithm of multi-direction saturation attack is built on reference route,route-flight-time,and impact azimuth.Simulation results show that the algorithm can realize missiles fired in a salvo launch reaching the target simultaneously from different directions while avoiding threat.

Iterative Learning Control for homing guidance design of missiles

This paper presents an Iterative Learning Control design applied to homing guidance of missiles against maneuvering targets. According to numerical experiments, although an increase of the control energies is appreciated with respect to a previous published base controller for comparison, this strategy, which is simple to realize, is able to reduce the time to reach the head-on condition to target destruction. This fact is important to minimize the missile lateral force-level to fulfill engaging in hyper-sonic target persecutions.

A distributed decision method for missiles autonomous formation based on potential game

The distributed cooperative decision problems of missiles autonomous formation with network packet loss are investigated by using the potential game based on formation principles.In particular,a dynamic target allocation method for missiles formation is provided based on the potential game and formation principles,after the introduction of cooperative guidance and control system of the missiles formation.Then we seek the optimization of a global utility function through autonomous missiles that are capable of making individually rational decisions to optimize their own utility functions.The first important aspect of the problem is to design an individual utility function considering the characteristics of the missiles formation,with which the objective of the missiles are localized to each missile yet aligned with the global utility function.The second is to equip the missiles with an appropriate coordination mechanism with each missile pursuing the optimization of its own utility function.We present the design procedure for the utility,and present a coordination mechanism based on spatial adaptive play and then introduce the idea of“cyclical selected spatial adaptive play”and“negotiation based on time division multiple address(TDMA)protocol formation support network”.Finally,we present simulations for the distributed dynamic target allocation on the comprehensive digital simulation system,and the results illustrate the effectiveness and engineering applicability of the method.

THE GLOBAL DEFORMATION AND LOCAL DAMAGE ANALYSIS OF FILLED METALLIC CYLINDRICAL SHELLS IMPACTED BY MISSILES

A group of theoretical models has been developed for analyzingtransient dynamic re- sponse to filled metallic cylindrical shellsimpacted by flat-nosed missiles at normal obliquity on the ba- sis ofthe analogy modeling method of beam-on-foundation. It can be used forsolving the global defor- mation and the local failure of cylindricalshells in the impact process. the ballistic limit speed and im- pactresponse history have been calculated by this group of theoreticalmodels, and the inner pressure effect on the ballistic limit speedand some parameters are discussed at length.

Target-tracked prioritization to surveille ballistic missiles

The missile-tracked priority assessment for the early warning system monitoring multi-missile very well, is the first task to defend them and useful to perform optimally the sensor-to-missile assignment. The problem of the missile-tracked priority assessment is of incomplete information, of multi-attribute and dynamic. To solve the dificult problem, the index system, which includes six classification indices, is established by means of reducing the target＇s primary information which the early warning system focuses on. The lack of some attributes values, which is caused by the incomplete information, is handled by the approach： first classifying each attribute as unknown one or known one, and then subdividing the latter, last using the expectation and the information entropy if the attribute is known but uncertain. With a view of reality, nine qualitative evaluation criteria are given. Based on them each index is quantified by the eigenvector method. And then based on the improved technique for order preference by similarity to ideal solution （TOPSIS）, the targets-tracked priorities are assessed and sorted by an evaluation function from three aspects： 1） the quantity of the available information, 2） the affirmative or accurate degree of the available information, 3） the classification or trait of the available information.

Optimal Guidance Law to Maximize Terminal Velocity for Missiles with Impact Angle Constraint

In this paper,an optimal guidance law for missiles with impact angle and miss distance constraints is proposed to achieve the maximal terminal velocity. The normal acceleration command that includes the timevarying coefficients is introduced to satisfy the desired impact angle as well as zero miss distance according to the geometric relation and relative motion parameters between missile and target. The problem is formulated as an optimal control problem by defining the angle of velocity error and flight-path angle as state variables and maximizing a performance index of the terminal velocity. The analytical form of the proposed guidance law is obtained as the solution of the optimal control problem combining optimal control theory and numerical value computation method. Nonlinear simulations of various situations demonstrate the performance and feasibility of the proposed optimal guidance law.

RUPTURE OF THIN METAL TUBES BY NORMAL AND OBLIQUE IMPACT OF BLUNT CONICALNOSED MISSILES:EXPERIMENTS

Impact tests at both normal and oblique angles of incidence were conducted on thin mild tubes using a moderate size of 90 degrees conical-nosed missiles. The minimum impact speed that generated cracks through the thickness of the wall, termed the speed for rupture, was measured, and various modes of rupture were identified. For a thin tube hit by a missile at a normal angle of obliquity at the speed for rupture, the contact region spreads across the nose of the missile, and the transverse shear deformation is predominant in the final failure process. If the angle of obliquity is 30 degrees, the missile pierces a hole through the wall of the tube. At the speed for rupture, the kinetic energy of the missile for oblique angle 30 degrees is only about 45% that required for plugging at a normal angle of obliquity.

Damage effectiveness assessment method for anti-ship missiles based on double hierarchy linguistic term sets and evidence theory

The research on the damage effectiveness assessment of anti-ship missiles involves system science and weapon science, and has essential strategic research significance. With comprehensive analysis of the specific process of the damage assessment process of anti-missile against ships, a synthetic damage effectiveness assessment process is proposed based on the double hierarchy linguistic term set and the evidence theory. In order to improve the accuracy of the expert ’s assessment information, double hierarchy linguistic terms are used to describe the assessment opinions of experts. In order to avoid the loss of experts ’ original information caused by information fusion rules, the evidence theory is used to fuse the assessment information of various experts on each case. Good stability of the assessment process can be reflected through sensitivity analysis, and the fluctuation of a certain parameter does not have an excessive influence on the assessment results. The assessment process is accurate enough to be reflected through comparative analysis and it has a good advantage in damage effectiveness assessment.

A METHOD IN SYSTEM DESIGN OF EJECTING DEVICES OF MISSILES

Anew method in system design of ejecting devices of missiles is first presented.Some important points are dis-cussed,which guid the research and development of new ejecting devices of missileg,amd provid the foundation flr thw design of mew ejecting device is provided.The system design includes the distribution of techmology specifica-tion,3-D solid modeling of ejecting devices of missiles im-ported from abroad,the design of pmeumatic device sys-tem,the design of ejecting mechanism system,the predic-tion of reliability and the experimental analysis,etc.

Evaluation and optimization of strapdown velocity numerical integration algorithms for SINS in spinning ballistic missiles

The error of the conventional velocity numerical integration algorithm was evaluated through the Taylor series expansion. It is revealed that neglecting the second- and higher-order terms of attitude increments will lead to the velocity numerical integration error, which is proportional to the triple cross product of the angular rate and specific force. A selection criterion for the velocity numerical integration algorithm was established for strapdown inertial navigation system (SINS) in spinning missiles. The spin angular rate with large amplitude will cause the accuracy of the conventional velocity numerical integration algorithm in SINS to decrease dramatically when the ballistic missile is spinning fast. Therefore, with the second- and higher-order terms of attitude increments considered, based on the rotation vector and the velocity translation vector, the velocity numerical integration algorithm was optimized for SINS in spinning ballistic missiles. The superiority of the optimized algorithm over the conventional one was analytically derived and validated by the simulation. The optimized algorithm turns out to be a better choice for SINS in spinning ballistic missiles and other high-precision navigation systems and high-maneuver applications.

Application of Fault Tolerant Control to Antiradiation Missiles

During the guidance process of an antiradiation missile (ARM), the target radar is often switched off in order to avoid being attacked. This will cause the passive seeker in ARM disabled, so the guidance system of the missile will stop working. To solve this problem, a scheme called LOS angles optimal estimation, has been put forward,based on the fault-tolerant design method of system reconstruction. The effectiveness of the proposed scheme has been proved by the simulations.

Task assignment in ground-to-air confrontation based on multiagent deep reinforcement learning

The scale of ground-to-air confrontation task assignments is large and needs to deal with many concurrent task assignments and random events.Aiming at the problems where existing task assignment methods are applied to ground-to-air confrontation,there is low efficiency in dealing with complex tasks,and there are interactive conflicts in multiagent systems.This study proposes a multiagent architecture based on a one-general agent with multiple narrow agents(OGMN)to reduce task assignment conflicts.Considering the slow speed of traditional dynamic task assignment algorithms,this paper proposes the proximal policy optimization for task assignment of general and narrow agents(PPOTAGNA)algorithm.The algorithm based on the idea of the optimal assignment strategy algorithm and combined with the training framework of deep reinforcement learning(DRL)adds a multihead attention mechanism and a stage reward mechanism to the bilateral band clipping PPO algorithm to solve the problem of low training efficiency.Finally,simulation experiments are carried out in the digital battlefield.The multiagent architecture based on OGMN combined with the PPO-TAGNA algorithm can obtain higher rewards faster and has a higher win ratio.By analyzing agent behavior,the efficiency,superiority and rationality of resource utilization of this method are verified.

Effect of direction error on the miss distance of rolling missiles with x-rudder

Based on the analysis of periodic equivalent control force of rolling missiles with x-rudder, the guidance loop model with direction error is established and the relationship between direction error and miss distance is analyzed. Results show that the miss distance is zero or a constant or infinite, and it is always zero when the real parts of system matrix eigenvalues decided by direction error are both positive values in an ideal system, in which all the lags are neglected. However, the miss distance gradually increases with the increase of the direction error and its variation is small when direction error is not more than 5° in the system, in which seeker lag and missile body lag are considered.

Research on terminal maneuver strategy of anti-ship missiles

To design the terminal maneuver strategy of an anti-ship missile,first,the analytical solution of miss distance when an anti-ship missile has planar weaving maneuver and three-dimension spiral maneuver is presented,in which not only the amplitude and frequency are considered but also the initial phase is taken into account.Next,based on the analytical solution of miss distance,the effects on the miss distance of the amplitude,frequency,initial phase of the anti-ship missile＇s maneuver acceleration and the order of flight control system of the air-ship missile are analyzed.Finally,the optimum weaving maneuver and spiral maneuver which make the miss distance be the largest under some conditions are designed,which is of important meaning for increasing the survival probability of the anti-ship missile.

Integrate Fuzzy Logic Stable Control System of Bank-to-Turn Missiles

The question of stable control system of bank-to-turn (BTT) missiles is a bottleneckin BTT technology. Integrate fuzzy logic stable control system of BTT missiles is designed in whichthree main problems are resolved. How to select input variables Of the fuzzy logic controller and howto guarantee completeness of the output control are two of them. The last one is how to coordinatethe fuzzy logic controllers in integrate fuzzy logic stable control system. Simulating results prov that integrate fuzzy logic stable coatrol system of BTT missiles is sueccessful, and it can be widelyused in future.