Collaborative perception leveraging intermediate feature fusion has emerged as a leading paradigm to significantly enhance the environmental perception capabilities of autonomous driving systems. However, existing methods typically rely on discriminative supervision guided by downstream tasks. This paradigm compels models to learn minimal, task-specific representations, which conflicts with the goal of cooperative perception to capture comprehensive information, thereby limiting generalization. To address this issue, we propose DiGS-CP, a novel two-stage generative supervised collaborative perception framework. Specifically, we introduce a diffusion-based generative task that conditions on fused object-level features to generate representations of object-level point clouds. The proposed generative supervision provides fine-grained, task-agnostic signals that encourages the fusion module to learn comprehensive representations beyond task-specific requirements. By preserving and integrating complementary information from collaborative agents, our approach overcomes the limitations of task-specific learning and enhances the generalizability of the learned features. Furthermore, our two-stage architecture requires agents to transmit only object-level features, significantly reducing communication overhead. Extensive experiments on three benchmark datasets demonstrate that DiGS-CP achieves state-of-the-art performance in 3D object detection, while maintaining low bandwidth requirements and exhibiting excellent generalization ability.

Heterogeneous collaborative perception enables interconnected agents to share and fuse intermediate features extracted from multimodal sensor data, thereby enhancing robustness and reducing perception blind spots. However, shared high-dimensional features often contain redundant and modality-specific information, leading to semantic inconsistencies and excessive communication overhead. To address these challenges, we propose an efficient communication framework for heterogeneous collaborative perception. This framework integrates a shared-private feature decoupling module that disentangles cross-modal features into shared and private components, sharing only shared feature to improve communication efficiency. Additionally, we design a spatial redundancy elimination module to remove background information from transmitted features, further reducing bandwidth consumption. Collectively, these modules enable compact and communication-efficient perception capabilities across heterogeneous modalities, models, and tasks. Experimental results demonstrate that the proposed framework significantly reduces communication overhead in complex heterogeneous scenarios while maintaining perception accuracy.

In the vehicular ad-hoc network (VANET) environment, the constant changes in traffic systems due to the real-time mobility of vehicles pose a challenge in developing communication strategies that can swiftly adapt to dynamic topological variations. Traditional protocols such as CSMA/CD offer dynamic access to shared media, but their efficiency is compromised by CSMA/CA's collision avoidance strategy, which reduces channel use due to its backoff algorithm. This paper introduces a novel Graph Convolutional Network-based Multi-Agent Reinforcement Learning MAC protocol (GM-MAC) for vehicular networks. Unlike the Neuro-DCF protocol, GMMAC incorporates local vehicular data directly, enabling more context-aware learning. Compared to other GNN-based MARL approaches like GraphComm, GMMAC better handles real-time changes and interactions among vehicles. Experimental results show that GMMAC significantly improves successful transmission numbers and reduces collisions, outperforming traditional CSMA/CA methods and other MARL approaches without GCN integration.