Search

Strategic minerals are the cornerstone of national security and stable development. China’s strategic mineral imports are highly dependent on maritime transport, and ensuring smooth maritime routes is crucial for safeguarding the security of strategic mineral imports. This study analyzes the overall pattern of China’s strategic mineral imports and maritime routes. Taking the iron ore maritime import network as an example, the study evaluates the importance and failure impacts of key maritime routes, summarizes the risk sources of important maritime routes, and proposes corresponding security strategies. The research finds that China’s strategic mineral imports are highly dependent on maritime routes. Based on the case study of iron ore maritime transport, the importance and failure impacts of maritime routes are identified. It is suggested to build differentiated and refined response strategies, focus on the protection of the Cape of Good Hope and the Lombok Strait transport corridor network, and pay attention to the safe navigation of large vessels. The risks of maritime routes for China’s strategic minerals mainly come from three aspects: navigational passages, international disputes and war threats, and non-traditional security issues. In response, multi-domain response strategies can be implemented, including building multiple strategic pivot nodes, constructing a multidimensional early warning system and an alternative route response mechanism, strengthening the construction of cross-border rail – sea intermodal transport corridors, conducting preventive diplomatic work, improving the maritime security system, and implementing hierarchical and differentiated risk-response measures.

The capability to predict catastrophic failures and anticipate challenges relating to trafficability enables quadruped robots to adjust their planned trajectory in time to prevent any damages. In this paper, an estimation‐based method is proposed to describe the validity of the trajectory from the viewpoint of the traction performance of the terrain. The main contribution of this study lies in equipping quadruped robots with the ability to comprehend the strength characteristics of the terrain and assess the validity of trajectory planning by anticipating potential catastrophic motion failures. To this end, a novel perception method of the foot terrain interface based on proprioception is proposed to estimate terrain strength parameters. The vehicle–terrain traction model is extended to a legged locomotion pattern to estimate the traction limits of the terrain. The trafficability anticipation is driven by combining the traction limits of the terrain and the expected ground reaction force, as obtained by the model‐based predictive methods. The validity of this method is verified using a quadruped robot test platform. Results highlight the potential of the presented approach, in which the anticipation of trafficability, grounded in the understanding of the terrain strength, serves as the explicit foundation for altering trajectories to avert motion failure. [ABSTRACT FROM AUTHOR]

The purpose of this study was to use the neural network to distinguish optic edema (ODE), and optic atrophy from normal fundus images and try to use visualization to explain the artificial intelligence methods. Three hundred and sixty-seven images of ODE, 206 images of optic atrophy, and 231 images of normal fundus were used, which were provided by two hospitals. A set of image preprocessing and data enhancement methods were created and a variety of different neural network models, such as VGG16, VGG19, Inception V3, and 50-layer Deep Residual Learning (ResNet50) were used. The accuracy, recall, F1-score, and ROC curve under different networks were analyzed to evaluate the performance of models. Besides, CAM (class activation mapping) was utilized to find the focus of neural network and visualization of neural network with feature fusion. Our image preprocessing and data enhancement method significantly improved the accuracy of model performance by about 10%. Among the networks, VGG16 had the best effect, as the accuracy of ODE, optic atrophy and normal fundus were 98, 90, and 95%, respectively. The macro-average and micro-average of VGG16 both reached 0.98. From CAM we can clearly find out that the focus area of the network is near the optic cup. From feature fusion images, we can find out the difference between the three types fundus images. Through image preprocessing, data enhancement, and neural network training, we applied artificial intelligence to identify ophthalmic diseases, acquired the focus area through CAM, and identified the difference between the three ophthalmic diseases through neural network middle layers visualization. With the help of assistant diagnosis, ophthalmologists can evaluate cases more precisely and more clearly. [ABSTRACT FROM AUTHOR]