In this paper, the distributed optimal formation control with collision avoidance for multiple QUAVs is studied. First, taking into account the limited sensing distance, the flight area is divided into three regions: detection region, avoidance region, and obstacle region. Then, a distributed observer is provided to estimate the states of the virtual leader. To handle collision avoidance, the more universal Lyapunov-like control barrier functions with a scheduling function are designed to modify the value function. By incorporating it into the fuzzy based reinforcement learning optimal control framework, it is proved that the trade-off between optimality and safety is realized. Finally, the performance of the framework is evaluated by the numerical simulation.

Three-dimensional (3D) polarized radiative transfer becomes increasingly important for remote sensing applications. In this study, we propose a 3D mesoscopic modeling for polarized radiative transfer in atmospheric systems based on the Lattice Boltzmann method (LBM). To accurately capture the complex geometries of 3D inhomogeneous atmospheres, the proposed LB model is constructed on unstructured grids. The accuracy of the proposed unstructured LB model is validated through comparison with benchmark solutions. Furthermore, this study employs the proposed unstructured LB model to investigate the influence of different infrared incident wavelengths and atmospheric characteristics on polarized radiative transfer. The results demonstrate that the unstructured LB model accurately captures the wavelength-dependent variations of the Stokes parameters in both forward and backward scattering processes, highlighting its exceptional capability in handling complex atmospheric structures and multiple optical components that influence polarized light. This model provides a 3D mesoscopic tool for remote sensing, particularly for interpreting polarized signals from inhomogeneous atmospheric systems.

Recently, word enhancement has become very popular for Chinese Named Entity Recognition (NER), reducing segmentation errors and increasing the semantic and boundary information of Chinese words. However, these methods tend to ignore the semantic relationship before and after the sentence after integrating lexical information. Therefore, the regularity of word length information has not been fully explored in various word-character fusion methods. In this work, we propose a Lexicon-Attention and Data-Augmentation (LADA) method for Chinese NER. We discuss the challenges of using existing methods in incorporating word information for NER and show how our proposed methods could be leveraged to overcome those challenges. LADA is based on a Transformer Encoder that utilizes lexicon to construct a directed graph and fuses word information through updating the optimal edge of the graph. Specially, we introduce the advanced data augmentation method to obtain the optimal representation for the NER task. Experimental results show that the augmentation done using LADA can considerably boost the performance of our NER system and achieve significantly better results than previous state-of-the-art methods and variant models in the literature on four publicly available NER datasets, namely Resume, MSRA, Weibo, and OntoNotes v4. We also observe better generalization and application to a real-world setting from LADA on multi-source complex entities.