To enhance guidance accuracy in near-space target interception, this paper presents an onboard deployment scheme of an attention-enhanced LSTM encoder network for real-time trajectory prediction. The system integrates Kalman filtering for motion reconstruction and maneuver recognition, enabling multi-modal prediction under complex target dynamics. A lightweight version of the model is deployed on a missile-borne ARM + Atlas200 + FPGA heterogeneous inference module, maintaining accuracy degradation within 10

This article is concerned with the periodic event-triggered (PET) control problem for spacecraft formation flying near asteroids. First, considering the limited sensing ability of the member spacecraft, an artificial potential function is incorporated to design a distributed error surface with connectivity preservation. Especially, by using the prescribed performance control technique, the convergence behavior of the proposed manifold can be actively adjusted. Then, a distributed event-triggered formation control law is designed based on the error surface. The controller inherits the properties of the proposed error variable, which would bring convenience to the theoretical analysis of the system performance. After that, the updating frequency of the proposed controller is further reduced by taking advantage of the PET strategy. Different from the continuous event-triggered mechanism, the triggering condition in the PET mechanism only needs to be evaluated by each spacecraft at the predetermined periodic sample instants, which is more feasible to be implemented onboard the spacecraft. Besides, the relationship between the state convergence region and the sampling period is analyzed. Finally, numerical examples are given to demonstrate the effectiveness of the proposed method. The simulation results show that the developed control strategy would enable the formation system to achieve the desired configuration with connectivity preservation, adjustable performance, and low-frequency control executions.

In this paper, the neural-network-based filtering problem is investigated for multi-sensor systems with dynamic encoding mechanisms. The sensor nodes and the remote filter are connected through bandwidth-constrained communication networks. To alleviate the communication burden, a novel dynamic encoding-based data compression-decompression mechanism is proposed so as to encode the data into a limited number of bits. Then, with the aid of the neural network learning method, a neural-network-based set-membership filter is developed for estimating the system states. Sufficient conditions are obtained to ensure that the filtering error remains within the bounded ellipsoidal set. In addition, the neural network tuning parameters and the filter gains are calculated by solving constrained optimization problems. Finally, the effectiveness of the proposed filtering algorithm is verified through a scenario of maneuvering target tracking using multiple unmanned aerial vehicles.

Active vibration suppression of spacecraft flexible appendages has always been a difficult problem in aerospace attitude maneuver missions. Facing this problem, the component synthesis vibration suppression (CSVS) method is proposed, which can suppress the vibration of the flexible attachment while the spacecraft maneuvers. However, this method has problems such as insufficient anti-interference ability. Therefore, this paper proposes a new control strategy, which combines CSVS with Linear Active Disturbance Rejection Control (LADRC) to further improve the performance of the CSVS method. Specifically, this paper first obtains the CSVS control torque according to the component synthesis design criterion, and applies it to the reference model to obtain the predetermined trajectory. Then, the LADRC controller is used to compensate the interference to ensure that the real trajectory coincides with the reference trajectory. Finally, the stability of the method is proved, and the effectiveness of the method is verified by simulation. The advantage of this method is that it can improve the vibration suppression performance of CSVS method in nonlinear attitude dynamics, and can effectively suppress the vibration amplitude of flexible appendages and improve the attitude control accuracy. Therefore, this control strategy has important practical significance for the control of flexible spacecraft.