This paper investigates the unmanned aerial vehicle scheduling problem (UAVSP) under gridded battlefield and logistics management. Compared with traditional scheduling problems that focus on task allocation and path planning, the battlefield UAVSP requires additional consideration of the military-specific factors of cargo conflict constraints (such as munitions versus flammable items, and food versus chemicals), i.e., different classes of materials cannot be transported by the same UAV at the same time. Therefore, in this paper, a mixed-integer linear programming (MILP) model is developed with the objective of minimizing the transportation cost, including transportation time and the number of UAVs used. Then, a two-stage self-learning Jaya algorithm (TSJA) is introduced, where the first stage dynamically records operation success rates to build a knowledge base, while the second stage leverages this knowledge to intelligently guide operation selection. Finally, three transportation strategy self-organization methods are proposed to reduce the transportation cost. In terms of solution quality, experimental results show that the TSJA outperforms the other algorithms.

The high-accuracy positioning of autonomous underwater vehicles (AUVs) is important in the Internet of Underwater Things (IoUT). AUV positioning relies on acoustic signal measurement systems. Owing to the variation in sound speed, the acoustic signal path is curved. The straight-line distance from an AUV to a measurement system cannot be simply calculated using the product of the propagation time and sound speed. Therefore, a time of arrival (TOA) positioning model is proposed to avoid calculating the straight-line distance, which is based on the continuous sound speed profile (SSP) ray tracing method. This model does not require knowing the depth of the AUV in advance. In addition, an improved Gauss-Newton iteration method based on the effective sound speed (ESS) is proposed to estimate the AUV position. The distance residual in the improved iteration is calculated on the basis of the time residual and the ESS, which is more accurate. Because the measured SSP is discrete data and cannot be directly used for the TOA positioning model, a Gauss kernel B-spline is proposed to obtain the continuous SSP. By optimizing the parameter σ, a high-accuracy continuous SSP can be obtained. Furthermore, theoretical analyses including the Cramér-Rao lower bound (CRLB) and complexity of the algorithm are also provided. The simulation results show that, compared with the 10th-degree polynomial and cubic standard B-spline, the Gauss kernel B-spline effectively improves the fitting accuracy. Compared with the traditional sound speed model and the current isogradient ray tracking model, the proposed TOA positioning model has better AUV positioning accuracy and is closer to CRLB.

In industrial practice, as systems become increasingly complex and integrated, the simultaneous failure of multicomponents, namely, coupling faults, has become more prevalent, which can be viewed as multilabel data. In addition, due to the changing industrial tasks, coupling fault diagnosis problems under varying operating conditions can be treated as cross-domain multilabel learning problems, which can be solved by multilabel transfer learning methods. However, existing multilabel transfer learning methods are all preliminary explorations lacking an in-depth exploring the multilevel similarity and complex features of coupling faults. To address this challenging problem, we propose a novel multilabel transfer learning method for coupling fault diagnosis, which achieves dual domain alignment at two levels. At the global feature level, the hypothesis space is reduced by minimizing the maximum mean discrepancy (MMD) at multistages of the network to align the global distribution. Furthermore, we decomposed the overall similarity into a combination of multiple local similarities and innovatively designed a dynamic multimetric structure to capture the diverse similarity characteristics of the data. By integrating the multimetric structure and dynamic mapping selection technique to form mathematical representations of this diverse similarity, this approach constrains the consistency of the local spatial structure of the two domains to achieve local space structure alignment. This method performs high superiority in multiple transfer tasks on the public and laboratory datasets, strongly demonstrating its effectiveness.