The formation control of unmanned aerial vehicle (UAV) swarms is of significant importance in various fields such as transportation, emergency management, and environmental monitoring. However, the complex dynamics, nonlinearity, uncertainty, and interaction among agents make it a challenging problem. In this paper, we propose a distributed robust control strategy that uses only local information of UAVs to improve the stability and robustness of the formation system in uncertain environments. We establish a nominal control strategy based on position relations and a semi-definite programming model to obtain control gains. Additionally, we propose a robust control strategy under the rotation set $\Omega$ to address the noise and disturbance in the system, ensuring that even when the rotation angles of the UAVs change, they still form a stable formation. Finally, we extend the proposed strategy to a quadrotor UAV system with high-order kinematic models and conduct simulation experiments to validate its effectiveness in resisting uncertain disturbances and achieving formation control.

The air-ground networked swarms have excellent application potential in national economics and production, such as smart cities, smart agriculture and forestry, and intelligent transportation. It also shows an exceptional value of application in the military fields, such as battlefield situational awareness and air-ground cooperative strikes. This paper addresses the need for accurate perception and recognition of complex environmental targets in air-ground networked swarms. We present a distributed learning and adaptive information fusion approach for air-ground network clustering, and we develop a model that minimizes the global likelihood function based on the probability of pattern categorization. This algorithm includes two main steps: information diffusion based on the gradient descent method and information fusion based on adaptive weighting calculation, and forms an air-ground cooperative pattern recognition method. Furthermore, the average error recursive equation for the cooperative pattern recognition algorithm of the air-ground networked clusters is derived, and the error convergence of the algorithm is theoretically proven. The results of the simulation demonstrate that the distributed fusion algorithm is accurate and that it can converge to the theoretical optimal level of the system in terms of both average mean square deviation and system error of information estimation. The simulation results show that the distributed fusion algorithm has good accuracy, and the algorithm's average mean square deviation and system error of information estimation can converge to the system's theoretical optimal level. When the number of nodes increases from 10 to 40, the mean square deviation of the distributed fusion algorithm for air-ground networked swarms decreases from −48.70 dB to −53.96 dB, and the system error reduces from −27.42 dB to −30.22 dB, which is close to the theoretical value of the error. Comparative experiments show that the algorithm proposed in this paper has good accuracy compared to traditional methods and can effectively support the perception and recognition of complex environmental targets for air-ground networked swarms.

Self-driving vehicles require a number of tests to prevent fatal accidents and ensure their appropriate operation in the physical world. However, conducting vehicle tests on the road is difficult because such tests are expensive and labor intensive. In this study, we used an autonomous-driving simulator, and investigated the three-dimensional environmental perception problem of the simulated system. Using the open-source CARLA simulator, we generated a CarlaSim from unreal traffic scenarios, comprising 15 000 camera-LiDAR (Light Detection and Ranging) samples with annotations and calibration files. Then, we developed Multi-Sensor Fusion Perception (MSFP) model for consuming two-modal data and detecting objects in the scenes. Furthermore, we conducted experiments on the KITTI and CarlaSim datasets; the results demonstrated the effectiveness of our proposed methods in terms of perception accuracy, inference efficiency, and generalization performance. The results of this study will faciliate the future development of autonomous-driving simulated tests.