The interception probability of a single missile is the basis for combat plan design and weapon performance evaluation,while its influencing factors are complex and mutually coupled.Existing calculation methods have v...The interception probability of a single missile is the basis for combat plan design and weapon performance evaluation,while its influencing factors are complex and mutually coupled.Existing calculation methods have very limited analysis of the influence mechanism of influencing factors,and none of them has analyzed the influence of the guidance law.This paper considers the influencing factors of both the interceptor and the target more comprehensively.Interceptor parameters include speed,guidance law,guidance error,fuze error,and fragment killing ability,while target performance includes speed,maneuverability,and vulnerability.In this paper,an interception model is established,Monte Carlo simulation is carried out,and the influence mechanism of each factor is analyzed based on the model and simulation results.Finally,this paper proposes a classification-regression neural network to quickly estimate the interception probability based on the value of influencing factors.The proposed method reduces the interference of invalid interception data to valid data,so its prediction accuracy is significantly better than that of pure regression neural networks.展开更多
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 u...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.展开更多
Hybrid sensor networks (HSNs) comprise of mobile and static sensor nodes setup for purpose of collaboratively performing tasks like sensing a phenomenon or monitoring a region. In this paper, we present target interce...Hybrid sensor networks (HSNs) comprise of mobile and static sensor nodes setup for purpose of collaboratively performing tasks like sensing a phenomenon or monitoring a region. In this paper, we present target interception as a novel application using mobile sensor nodes as executor. Static sensor nodes sense, compute and communicate with each other for navigation. Mobile nodes are guided to intercept target by the static nodes nearby. Our approach does not require any prior maps of the environment thus, cutting down the cost of the overall energy consumption. As to multi-targets multi-mobile nodes case, we present a PMB algorithm for task assignment. Simulation results have verified the feasibility and effectiveness of our approach proposed.展开更多
This paper investigates the guidance method based on reinforcement learning(RL)for the coplanar orbital interception in a continuous low-thrust scenario.The problem is formulated into a Markov decision process(MDP)mod...This paper investigates the guidance method based on reinforcement learning(RL)for the coplanar orbital interception in a continuous low-thrust scenario.The problem is formulated into a Markov decision process(MDP)model,then a welldesigned RL algorithm,experience based deep deterministic policy gradient(EBDDPG),is proposed to solve it.By taking the advantage of prior information generated through the optimal control model,the proposed algorithm not only resolves the convergence problem of the common RL algorithm,but also successfully trains an efficient deep neural network(DNN)controller for the chaser spacecraft to generate the control sequence.Numerical simulation results show that the proposed algorithm is feasible and the trained DNN controller significantly improves the efficiency over traditional optimization methods by roughly two orders of magnitude.展开更多
基金supported by the Foundation Strengthening Program Technology Field Foundation(2020-JCJQ-JJ-132)。
文摘The interception probability of a single missile is the basis for combat plan design and weapon performance evaluation,while its influencing factors are complex and mutually coupled.Existing calculation methods have very limited analysis of the influence mechanism of influencing factors,and none of them has analyzed the influence of the guidance law.This paper considers the influencing factors of both the interceptor and the target more comprehensively.Interceptor parameters include speed,guidance law,guidance error,fuze error,and fragment killing ability,while target performance includes speed,maneuverability,and vulnerability.In this paper,an interception model is established,Monte Carlo simulation is carried out,and the influence mechanism of each factor is analyzed based on the model and simulation results.Finally,this paper proposes a classification-regression neural network to quickly estimate the interception probability based on the value of influencing factors.The proposed method reduces the interference of invalid interception data to valid data,so its prediction accuracy is significantly better than that of pure regression neural networks.
基金supported by the National Natural Science Foundation of China(Grant No.12072090)。
文摘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.
文摘Hybrid sensor networks (HSNs) comprise of mobile and static sensor nodes setup for purpose of collaboratively performing tasks like sensing a phenomenon or monitoring a region. In this paper, we present target interception as a novel application using mobile sensor nodes as executor. Static sensor nodes sense, compute and communicate with each other for navigation. Mobile nodes are guided to intercept target by the static nodes nearby. Our approach does not require any prior maps of the environment thus, cutting down the cost of the overall energy consumption. As to multi-targets multi-mobile nodes case, we present a PMB algorithm for task assignment. Simulation results have verified the feasibility and effectiveness of our approach proposed.
基金supported by the National Defense Science and Technology Innovation(18-163-15-LZ-001-004-13).
文摘This paper investigates the guidance method based on reinforcement learning(RL)for the coplanar orbital interception in a continuous low-thrust scenario.The problem is formulated into a Markov decision process(MDP)model,then a welldesigned RL algorithm,experience based deep deterministic policy gradient(EBDDPG),is proposed to solve it.By taking the advantage of prior information generated through the optimal control model,the proposed algorithm not only resolves the convergence problem of the common RL algorithm,but also successfully trains an efficient deep neural network(DNN)controller for the chaser spacecraft to generate the control sequence.Numerical simulation results show that the proposed algorithm is feasible and the trained DNN controller significantly improves the efficiency over traditional optimization methods by roughly two orders of magnitude.
