This paper addresses the critical challenge of maintaining platoon stability amidst random communication delays and dynamic traffic scenarios. Existing strategies often face a trade-off between adaptability and theoretical guarantees. Simple controllers lack robustness to network uncertainties, while complex data-driven methods may lack formal stability proofs. To bridge this gap, we propose a Reliability-Aware and Attention-Driven Distributed Model Predictive Control (RAAD-DMPC) framework. The strategy models vehicle-to-everything (V2X) network reliability and employs an attention mechanism to dynamically fuse reference trajectories by weighting real-time link quality and delay. Based on this, a nonlinear MPC controller optimizes each vehicle’s input, with guaranteed theoretical boundedness via Input-to-State Stability (ISS) analysis. Extensive simulations demonstrate that RAAD-DMPC significantly enhances tracking accuracy, comfort, and fuel efficiency. Compared to the NRV2X-PID baseline, performance improvements of up to 96.2%, 85.3%, and 59.7% are achieved in these respective metrics. The framework’s scalability and robustness are further validated in large-scale platoons under severe delays, showcasing its potential as a robust and communication-aware framework for future cooperative vehicle systems.

Accurately identifying drug-target interactions (DTIs) is crucial for accelerating drug discovery. While computational methods offer efficient alternatives to experimental validation, most existing models struggle to integrate diverse structural and sequential characteristics of drugs and proteins within a unified framework. We propose UMSDTI, a deep learning model that captures molecular structural information from sequence, atom, and bond perspectives, and encodes proteins via integrating sequence and pre-trained modelderived contact maps. A co-attention fusion module jointly learns drug-target dependencies for enhanced prediction. Experiments on five benchmark datasets demonstrate that UMSDTI outperforms state-of-the-art methods in both prediction performance and biological insight. Source codes are available at https://anonymous.4open.science/r/UMSDTI.

This paper presents a distributed platoon control method designed to mitigate the impact of dynamic network interruptions on vehicle platoon safety and stability. Such interruptions pose significant challenges to maintaining platoon formation and achieving effective tracking performance. The proposed approach integrates network performance and state estimation within a nonlinear distributed model predictive control (DMPC) framework. A V2V communication model for platooning on straight roads is developed using stochastic geometry theory, and a unidirectional connectivity determination method assesses platoon topology connectivity. Missing information due to communication degradation is estimated using the probabilities of successful V2X information reception and connectivity results. This estimation provides reference trajectories for disconnected vehicles, ensuring platoon safety and stability. The performance of the proposed method is evaluated through simulations of a heterogeneous platoon with 15 vehicles under nine network topologies, considering both fixed-time and random intermittent network interruptions. Results demonstrate that under random network interruptions, the proposed method improves tracking performance by 87.1%, driving comfort by 72.5%, and energy efficiency by 23.2% compared to a baseline control method. Preliminary real-world testing further validates the feasibility of the approach. These findings underscore the effectiveness of the proposed method in maintaining platoon stability and safety, even during complete V2X communication loss.

The goal of multi-object tracking (MOT) is to estimate the location of objects and maintain their identities consistently to yield their individual trajectories. It has become a trend to fuse multi-sensor information to achieve 3D MOT, since it can leverage the advantages of different sensors to enhance tracking performance. However, it is a challenging work due to the necessity of fusing features with diverse attributes and wrong association caused by significant noise. In this paper, we propose adaptive weight parameter-based multi-feature fusion to create affinity function, alongside adaptive setting of data association searching range, aiming to augment camera-Lidar information fusion-based MOT framework. First, detected results from these two sensors are divided into three categories. Then, to fully utilize both motion and appearance information, adaptive weight parameter setting is proposed to embed appearance information into motion information, forming the basis for creating an affinity function for data association. Furthermore, to mitigate wrong association caused by object temporary occlusion or out-of-view, a method for adaptively adjusting the association searching area is introduced based on the number of frames in which tracking trajectories disappear. Finally, to prevent appearance information pollution caused by significant noise, a confidence score-based tracking trajectory appearance feature updating strategy is explored. The experiment results on KITTI and nuScenes MOT benchmark show remarkable performance improvement over other state-of-the-art MOT methods and demonstrate the effect of our designed modules.

Feature pyramid network (FPN) can highly improve the performance of object detection by extracting multiscale features. However, current FPN-based methods suffer from intrinsic correlation of local information loss in each feature map, which brings about the semantic information effective transmission problem. In addition, 1 × 1 convolution in lateral connection of FPN may cause spatial information loss. In this article, we propose a novel semantic and spatial information enhancing feature pyramid network (Dual SIE-FPN), which mainly focuses on alleviating multiscale hierarchical feature transmission loss and enhancing the feature representation. Specifically, Dual SIE-FPN contains three modules: Lateral Feature Enhancement (LFE), Global Attention Upsampling (GAU), and Multiple Information Compensation (MIC). LFE is designed to capture deep semantic representation and enhance channel information. GAU is established to make up for spatial information loss caused by upsampling, and transmit the high-level features with the compensatory information to low-level features simultaneously. MIC is designed to work with LFE in parallel to further improve the information loss resulting from 1 × 1 convolution. Experimental results on MS COCO and UAVDT dataset demonstrate that Dual SIE-FPN achieves competitive performance compared to other state-of-the-art FPNs. In addition, our proposed Dual SIE-FPN can be embedded into any multiscale feature extraction-based computer vision tasks to improve the performance.

In vehicular ad hoc networks, the exchange of safety messages plays a crucial role in transportation efficiency and safety. However, safety messages are generally only signed rather than encrypted when broadcast, which lures malicious attackers to collect sensitive and private data. Obviously, an efficient privacy-preserving signcryption mechanism is essential for the timely and correct acceptance of safety messages. Currently, most state-of-the-art mechanisms satisfy security requirements, but performance is inefficient. Therefore, in this article, we design a provably secure lightweight authentication based on elliptic curve signcryption (EPSLA) for vehicle-to-vehicle communication, which can fulfill batch unsigncryption and a feasible security communication scheme in practical scenarios. The EPSLA scheme ensures security in relation to existential unforgeability against adaptive chosen message attacks and indistinguishability against adaptive chosen ciphertext attacks under the hardness assumption of the elliptic curve discrete logarithm problem in the random oracle model. Performance analysis shows that our scheme significantly performs better than the comparison schemes.