面向车队云协同计算的深度神经网络推理任务拆分与卸载策略研究OA

With the wide deployment of deep neural networks(DNNs)in autonomous driving perception tasks such as object tracking and semantic segmentation,the contradiction between stringent real-time inference requirements and limited onboard computing resources has become a major bottleneck restricting system performance.To address this is-sue,this paper proposes a DNN task partitioning and offloading strategy for platooning cloud environments.First,a dual-dimensional collaborative partitioning and offloading model is developed with the objective of minimizing infer-ence latency.In the inter-layer dimension,DNN tasks are adaptively partitioned according to heterogeneous vehicle computing capabilities and link conditions.In the intra-layer dimension,the parts with heavy single-layer computational loads are further spatially partitioned and assigned to multiple vehicles for parallel execution,thereby alleviating local computational bottlenecks.Second,the joint optimization problem of dynamic task partitioning and offloading is formu-lated as a decentralized partially observable Markov decision process(Dec-POMDP),and a multi-agent deep determin-istic policy gradient(MADDPG)based solution is designed to achieve dynamic decision optimization in platooning cloud environments.Simulation results show that the proposed method can effectively reduce DNN inference latency under different platoon scales,computing capacities,and bandwidth settings.Compared with full local computing,whole-model offloading,and conventional inter-layer partitioning schemes,the proposed method achieves superior latency performance,especially under resource-constrained conditions.