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
The controller design and digital simulation for the hyper velocity kinetic energy missile is investigated. A mathematical model of the trajectory deviation from the line of sight was established, the guidance closed ...The controller design and digital simulation for the hyper velocity kinetic energy missile is investigated. A mathematical model of the trajectory deviation from the line of sight was established, the guidance closed loop was compensated with a phase advance lag corrective network, a selecting algorithm of the attitude control motors used to steer the missile's attitude was presented. In the presence of a wide variety of disturbances the results of digital simulation are satisfactory to circular error probability(CEP) being less than 0 5?m. The steering scheme utilizing attitude control motors as actuators to control the attitude of the missile is feasible.展开更多
In order to increase the aircraft＇s survival in the flight mission, it is necessary to carry out flight mission planning, which includes TF/TA2（ terrain following/terrain avoidance/threat avoidance）. An approach to...In order to increase the aircraft＇s survival in the flight mission, it is necessary to carry out flight mission planning, which includes TF/TA2（ terrain following/terrain avoidance/threat avoidance）. An approach to 3D-route planning based on A ＊ heuristic search algorithm was selected to determine the routes of fiber optic guidance missile＇ s cruise segment. The cost function was discussed, which was mainly related to the physical obstacle, threat exposure, and aircraft performance characteristics. The digital map techniques were presented, which included setting ＂no-go＂ area according to fiber＇s safety requirement. The optimal or the sub-optimal route was obtained, while the cost function constraints were satisfied and the stored terrain obtained from a real terrain was digitized. The algorithm is validated through simulation and can fulfill the route planning task which focuses on the cruise segment of fiber optic guidance missile.展开更多
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The controller design and digital simulation for the hyper velocity kinetic energy missile is investigated. A mathematical model of the trajectory deviation from the line of sight was established, the guidance closed loop was compensated with a phase advance lag corrective network, a selecting algorithm of the attitude control motors used to steer the missile's attitude was presented. In the presence of a wide variety of disturbances the results of digital simulation are satisfactory to circular error probability(CEP) being less than 0 5?m. The steering scheme utilizing attitude control motors as actuators to control the attitude of the missile is feasible.
基金Sponsored by the Ministerial Level Advanced Research Foundation(51401050105BQ01)
文摘In order to increase the aircraft＇s survival in the flight mission, it is necessary to carry out flight mission planning, which includes TF/TA2（ terrain following/terrain avoidance/threat avoidance）. An approach to 3D-route planning based on A ＊ heuristic search algorithm was selected to determine the routes of fiber optic guidance missile＇ s cruise segment. The cost function was discussed, which was mainly related to the physical obstacle, threat exposure, and aircraft performance characteristics. The digital map techniques were presented, which included setting ＂no-go＂ area according to fiber＇s safety requirement. The optimal or the sub-optimal route was obtained, while the cost function constraints were satisfied and the stored terrain obtained from a real terrain was digitized. The algorithm is validated through simulation and can fulfill the route planning task which focuses on the cruise segment of fiber optic guidance missile.
