The task of vision-and-language navigation in continuous environments (VLN-CE) aims at training an autonomous agent to perform low-level actions to navigate through 3D continuous surroundings using visual observations and language instructions. The significant potential of VLN-CE for mobile robots has been demonstrated across a large number of studies. However, most existing works in VLN-CE focus primarily on transferring the standard discrete vision-and-language navigation (VLN) methods to continuous environments, overlooking the problem of collisions. Such oversight often results in the agent deviating from the planned path or, in severe instances, the agent being trapped in obstacle areas and failing the navigational task. To address the above-mentioned issues, this letter investigates various collision scenarios within VLN-CE and proposes a classification method to predicate the underlying causes of collisions. Furthermore, a new VLN-CE algorithm, named Safe-VLN, is proposed to bolster collision avoidance capabilities including two key components, i.e., a waypoint predictor and a navigator. In particular, the waypoint predictor leverages a simulated 2D LiDAR occupancy mask to prevent the predicted waypoints from being situated in obstacle-ridden areas. The navigator, on the other hand, employs the strategy of ‘re-selection after collision’ to prevent the robot agent from becoming ensnared in a cycle of perpetual collisions. The proposed Safe-VLN is evaluated on the R2R-CE, the results of which demonstrate an enhanced navigational performance and a statistically significant reduction in collision incidences.

Small object detection has always been a difficult direction in the field of object detection, especially the detection of small objects in UAV aerial images. The images captured by UAVs have the characteristics of small objects and dense objects. In order to solve these two problems, this paper improves the performance of object detection from the aspects of data and network structure. In terms of data, the data augmentation strategy and image pyramid mechanism are mainly used. The data augmentation strategy adopts the method of image division, which can greatly increase the number of small objects, making it easier for the algorithm to be fully trained during the training process. Since the object is denser, the image pyramid mechanism is used. During the training process, the divided images are up-sampled into three different sizes, and then sent to three different detectors respectively. Finally, the detection results of the three detectors are fused to obtain the final detection results. The small object itself has few pixels and few features. In order to improve the detection performance, it is necessary to use context. This paper adds attention mechanism to the yolov5 network structure, while adding a detection head to the underlying feature map to make the network structure pay more attention to small objects. By using data augmentation and improved network structure, the detection performance of small objects can be significantly improved. The experiment in this paper is carried out on the Visdrone2019 dataset and DOTA dataset. Through experimental verification, our proposed method can significantly improve the performance of small object detection.

Enhanced by onboard computing resources and inter-satellite links (ISLs), large-scale satellite networks (SNs) can supplement terrestrial networks, facilitating real-time edge computing and transmission in remote areas at low latency. This capability benefits applications such as real-time remote object tracking, vehicular edge computing, and environmental monitoring. However, different from terrestrial vehicular networks, satellite mobility and the intermittent nature of ISLs often render many computing satellites inaccessible, necessitating efficient scheduling scheme for joint computing and communication resource allocation. To meet this requirement, we investigate joint computing and transmission strategy for real-time computing applications over SNs to reduce end-to-end delays. We formulate the real-time computing and transmission problem over time-varying SNs as an integer linear programming problem. The mathematical optimization method has an exponential time complexity of ${O}(2|\mathcal {L}^{\tau }| \cdot (4 |\mathcal {V}^{\tau }| + 4 |\mathcal {L}^{\tau }|)^{2|\mathcal {L}^{\tau }|})$ in the worst case, which grows rapidly with the growth of network sizes. Here, $|\mathcal {V}^{\tau }|$ and $|\mathcal {L}^{\tau }|$ denote the number of nodes and links, respectively, within the time window ${\tau }$. Resorting to the designed $k$-shortest path-based method yields a worst-case complexity of $O(|\mathcal {V}^{\tau }|! \cdot |\mathcal {V}^{\tau }|^{3})$. While efficient under common scenarios, its running time is still relatively unstable and may become tens of times longer when computing resources become scarce. To enhance efficiency, we further design a graph-based algorithm that leverages the solution space's unique structure, thereby achieving optimal solutions with polynomial time complexity ${O}(3|\mathcal {L}^{\tau }| + (2\log {|\mathcal {V}^{\tau }|}+1) |\mathcal {V}^{\tau }|)$. Extensive evaluations conducted on the world's largest Starlink system validate the efficacy of the designed methods and highlight their potential synergy.

Cross-docking is a very useful logistics technique that can substantially reduce distribution costs and improve customer satisfaction. A key problem in its success is truck scheduling, namely, decision on assignment and docking sequence of inbound/outbound trucks to receiving/shipping dock doors. This paper focuses on the problem with the requirement of unloading/loading products in a given order, which is very common in many industries, but is less concerned by existing researches. An integer programming model is established to minimise the makespan. An improved particle swarm optimisation (ωc-PSO) algorithm is proposed for solving it. In the algorithm, a cosine decreasing strategy of inertia weight is designed to dynamically balance global and local search. A repair strategy is put forward for continuous search in the feasible solution space and a crossover strategy is presented to prevent the algorithm from falling into local optimum. After algorithm parameters are tuned using Taguchi method, computational experiments are conducted on different problem scales to evaluate ωc-PSO against genetic algorithm, basic PSO and GLNPSO. The results show that ωc-PSO outperforms other three algorithms, especially when the number of dock doors, trucks and product types is great. Statistical tests show that the performance difference is statistically significant.

In terms of enhancing the spectrum-sharing capability of the Internet of vehicles (IoV), the integrated sensing and communication (ISAC) systems of the communication transmission and radar detection in the IoV are discussed. Firstly, the optimal power allocation of communication and radar is considered, respectively, and the joint optimization problem of maximizing the communication rate and the Fisher information (FI) of radar sensing is constructed. Then, an adaptive optimization weight factor is introduced to optimize the power allocation of the ISAC system, to achieve a trade-off performance between sensing and communication in the IoV system. Subsequently, the alternating optimization fractional programming and KKT (AO-FP-KKT) algorithm is proposed based on the coupled characteristics of the problem. This algorithm introduces dual variables to construct the Lagrange function, combines fractional programming architecture, and utilizes KKT conditions to obtain closed-form solutions. In particular, the scope of the dual variable is analyzed in detail and proved strictly. Finally, the numerical simulation results show that the effectiveness of the proposed algorithm and its superior performance compared with the existing state-of-the-art power allocation methods are demonstrated. The proposed algorithm enhances the system spectrum-sharing capability and achieves a trade-off between sensing and communication performance.