This article aims to address the fixed-time optimal leader-following consensus issue for unknown multiagent systems (MASs) under Denial of Service (DoS) and false data injection (FDI) attacks. A novel fixed-time stability theorem under DoS attacks is presented to simplify the stability conditions and decrease the computational complexity of the settling time. Simultaneously, the deep neural networks (DNNs) structure with the projection operator are adopted in real-time to approximate the unknown system dynamics. To achieve the optimal consensus under cyber-attacks, a hierarchical control approach is presented, which includes a reference signal generation layer and a tracking control layer. Specifically, the distributed and Luenberger-based observers are designed in the reference signal generation layer to solve the fixed-time state estimation issues of leader and followers under multiple malicious attacks, respectively. Then, the optimal control strategy based on the event-triggered mechanism (ETM) is designed in the tracking control layer to track the reference signal and minimize the cost consumption. Due to the difficulty in obtaining explicit expressions of the optimal control mechanisms, a critic-only reinforcement learning (RL)-based algorithm is presented for online learning the unknown weight within a fixed time. By rigorous proof, the developed observers can achieve the fixed-time state reconstruction and the optimal control policy can track observation states after a fixed time. Finally, simulation results about platooning control of automated vehicles are given to demonstrate the efficacy of the developed technique.

Small target features are difficult to distinguish and identify in an environment with complex backgrounds. The identification and extraction of multi-dimensional features have been realized due to the rapid development of deep learning, but there are still redundant relationships between features, reducing feature recognition accuracy. The YOLOv5 neural network is used in this paper to achieve preliminary feature extraction, and the minimum redundancy maximum relevance algorithm is used for the 512 candidate features extracted in the fully connected layer to perform de-redundancy processing on the features with high correlation, reducing the dimension of the feature set and making small target feature recognition a reality. Simultaneously, by pre-processing the image, the feature recognition of the pre-processed image can be improved. Simultaneously, by pre-processing the image, the feature recognition of the pre-processed image can significantly improve the recognition accuracy. The experimental results demonstrate that using the minimum redundancy maximum relevance algorithm can effectively reduce the feature dimension and identify small target features.

This article focuses on the leader–following consensus in fixed time for multiagent systems under denial-of-service (DoS) attacks. First, the node-based distributed DoS attack model is developed, where multiple adversaries launch different attack strategies working together on nodes of the system. Second, a novel fixed-time stability theorem is presented, where an index function is introduced to decrease the complexity of the stability condition and achieve the stability goal under node-based distributed DoS attacks. Third, based on this index function, the resilient dynamic event-triggered mechanism (DETM) is designed by utilizing the DoS attacks detection signals to reduce the consumption of computational costs and save the system resources. With the assistance of Lyapunov stability theory, some consensus conditions are established and the Zeno behavior is excluded. Finally, the autonomous swarm systems, including autonomous aerial vehicles and autonomous ground vehicles, are applied to illustrate the effectiveness of control strategy.

Unmanned Aerial Vehicle (UAV) path planning research refers to the UAV automatically planning an optimal path to the destination under the corresponding environment, while avoiding collision with obstacles in this process. In order to solve the problem of 3D path planning of UAV in a dynamic environment, a heuristic dynamic reward function is designed to guide the UAV. We propose the Environment Exploration Twin Delayed Deep Deterministic Policy Gradient (EE-TD3) algorithm, which combines the symmetrical 3D environment exploration coding mechanism on the basis of TD3 algorithm. The EE-TD3 algorithm model can effectively avoid collisions, improve the training efficiency, and achieve faster convergence speed. Finally, the performance of the EE-TD3 algorithm and other deep reinforcement learning algorithms was tested in the simulation environment. The results show that the EE-TD3 algorithm is better than other algorithms in solving the 3D path planning problem of UAV.

Vehicle dynamics model-aided inertial navigation system (INS) is an effective approach for autonomous navigation when the Global Navigation Satellite System fails, However, the accuracy of the vehicle dynamics model (VDM) will decrease, or even diverge, in low-speed and high-dynamic situations, which leads to the performance degradation of the whole system. The traditional VDM-aided INS uses the velocity information from VDM to estimate and correct INS errors, but most previous research ignored that the acceleration and angular velocity from INS are more accurate than VDM. To reach the full potential of INS and VDM, this paper proposed a three-layer structure for the VDM-aided INS. In the proposed method, the navigation information is divided into the sensor layer, system layer I, and system layer II. In the sensor layer, VDM is corrected by INS through the strong tracking cubature Kalman filter, and the chi-square test is adopted to avoid redundant computation. In the system layer I, INS is aided by the corrected velocity information from VDM. In the system layer II, the position is updated with the velocity and attitude that are corrected in the system layer I. As the result, VDM and INS can be corrected in different layers simultaneously, and the accuracy and robustness of the whole system can be improved. The simulation experiment proved that, with the sensor layer filter, the velocity errors of VDM can be reduced by 35% in low-speed and high-dynamic situations, and reduced by 20% in the normal situation. Then, a 31-minute field test shows that the maximum horizontal position error of the proposed method is 241.8 m, while those of two traditional VDM-aided INS methods are more than 500 m.