[1]DENG Yixiang,ZHANG Yulin,BAI Long.Task partitioning and offloading strategy for DNN inference in platooning cloud[J].CAAI transactions on intelligent systems,2026,21(5):1260-1267.[doi:10.11992/tis.202512029]
Copy

Task partitioning and offloading strategy for DNN inference in platooning cloud

CAAI transactions on intelligent systems[ISSN 1673-4785/CN 23-1538/TP] Volume: 21 Number of periods: 2026 5 Page number: 1260-1267 Column: 学术论文—机器感知与模式识别 Public date: 2026-09-05
Title:
Task partitioning and offloading strategy for DNN inference in platooning cloud
Author(s):
DENG Yixiang1;  ZHANG Yulin1;  BAI Long2
1. College of Electrical Engineering and intelligent manufacturing, Chongqing metropolitan College of Science and Technology, Chongqing 402160, China;
2. College of Mechanical and Vehicle Engineering, Chongqing University, Chongqing 400044, China
Keywords:
deep neural networks; autonomous driving; platooning cloud; task partitioning; computation offloading; dual-dimensional splitting; multi-agent reinforcement learning; resource allocation
CLC:
TP183
DOI:
10.11992/tis.202512029
Abstract:
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 issue, 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 inference 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 formulated as a decentralized partially observable Markov decision process (Dec-POMDP), and a multi-agent deep deterministic 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.
References:
[1] 李克潮, 张继成. 多阶特征向量融合的深度神经网络产品推荐算法[J]. 计算机工程与设计, 2025, 46(6): 1742-1749 LI Kechao, ZHANG Jicheng. Deep neural network product recommendation algorithm based on multi-level feature vector fusion[J]. Computer engineering and design, 2025, 46(6): 1742-1749
[2] XU Yuanyuan, ZHU Kun, XU Hu, et al. Deep reinforcement learning for multi-objective resource allocation in multi-platoon cooperative vehicular networks[J]. IEEE transactions on wireless communications, 2023, 22(9): 6185-6198
[3] 杨孟璋, 农丽萍, 李然, 等. 基于一维卷积神经网络的雷达个体识别算法[J]. 计算机工程与设计, 2025, 46(5): 1281-1288 YANG Mengzhang, NONG Liping, LI Ran, et al. Radar individual recognition algorithm based on one-dimensional convolutional neural networks[J]. Computer engineering and design, 2025, 46(5): 1281-1288
[4] 徐钰龙, 李君, 李正权, 等. 基于深度强化学习的多用户蜂窝网络能效优化[J]. 计算机工程与设计, 2025, 46(3): 734-740 XU Yulong, LI Jun, LI Zhengquan, et al. Energy efficiency optimization of multi-user cellular networks based on deep reinforcement learning[J]. Computer engineering and design, 2025, 46(3): 734-740
[5] 于晶, 鲁凌云, 李翔. 车联网中基于DDQN的边云协作任务卸载机制[J]. 计算机工程, 2022, 48(12): 156-164 YU Jing, LU Lingyun, LI Xiang. Edge-cloud collaborative task offloading mechanism based on DDQN in vehicular networks[J]. Computer engineering, 2022, 48(12): 156-164
[6] 刘先锋, 梁赛, 李强, 等. 基于深度强化学习的云边协同DNN推理[J]. 计算机工程, 2022, 48(11): 30-38 LIU Xianfeng, LIANG Sai, LI Qiang, et al. Cloud-edge collaborative DNN inference based on deep reinforcement learning[J]. Computer engineering, 2022, 48(11): 30-38
[7] 咸琳涛, 刘晓兰, 王淦, 等. 异构环境感知的分布式神经网络训练模型[J]. 计算机工程与设计, 2024, 45(9): 2821-2827 XIAN Lintao, LIU Xiaolan, WANG Gan, et al. Distributed neural network training model for heterogeneous environment perception[J]. Computer engineering and design, 2024, 45(9): 2821-2827
[8] 赵婵婵, 吕飞, 石宝, 等. 面向边缘智能的协同推理方法研究综述[J]. 计算机工程与应用, 2025, 61(3): 1-20 ZHAO Chanchan, L? Fei, SHI Bao, et al. Review of collaborative inference methods for edge intelligence[J]. Computer engineering and applications, 2025, 61(3): 1-20
[9] SHAO Jiawei, ZHANG Jun. Communication-computation trade-off in resource-constrained edge inference[J]. IEEE communications magazine, 2020, 58(12): 20-26
[10] XIAO Yilin, XIAO Liang, WAN Kunpeng, et al. Reinforcement learning based energy-efficient collaborative inference for mobile edge computing[J]. IEEE transactions on communications, 2023, 71(2): 864-876
[11] 周燕, 许业文, 蒲磊, 等. 自动驾驶场景下的图像三维目标检测研究进展[J]. 计算机科学, 2024, 51(11): 133-147 ZHOU Yan, XU Yewen, PU Lei, et al. Research progress of image 3D object detection in autonomous driving scenario[J]. Computer science, 2024, 51(11): 133-147
[12] LI Xian, BI Suzhi. Optimal AI model splitting and resource allocation for device-edge co-inference in multi-user wireless sensing systems[J]. IEEE transactions on wireless communications, 2024, 23(9): 11094-11108
[13] ZENG Liekang, CHEN Xu, ZHOU Zhi, et al. CoEdge: cooperative DNN inference with adaptive workload partitioning over heterogeneous edge devices[J]. IEEE/ACM transactions on networking, 2021, 29(2): 595-608
[14] 王奕峰, 周婷, 徐天衡. 一种NOMA系统中的反馈深度神经网络检测算法[J]. 电讯技术, 2023, 63(5): 611-617 WANG Yifeng, ZHOU Ting, XU Tianheng. A feedback deep neural network detection algorithm in NOMA systems[J]. Telecommunication engineering, 2023, 63(5): 611-617
[15] BOUTIBA K, BAGAA M, KSENTINI A. Multi-agent deep reinforcement learning to enable dynamic TDD in a multi-cell environment[J]. IEEE transactions on mobile computing, 2024, 23(5): 6163-6177
[16] 刘锋, 张爽, 黄渝昂. 融合压缩采样与深度神经网络的直接序列扩频参数估计[J]. 电讯技术, 2022, 62(9): 1248-1253 LIU Feng, ZHANG Shuang, HUANG Yuang. Estimation of DSSS signal parameter based on compressive sampling and deep neural networks[J]. Telecommunication engineering, 2022, 62(9): 1248-1253
[17] 鲁富荣, 原之安, 钱宇华. 融合模体图神经网络和自编码器的链路预测[J]. 计算机科学与探索, 2023, 17(1): 209-216 LU Furong, YUAN Zhian, QIAN Yuhua. Link prediction combining motif graph neural network and auto-encoder[J]. Journal of frontiers of computer science & technology, 2023, 17(1): 209-216
[18] WANG Song, ZHANG Xinyu, UCHIYAMA H, et al. HiveMind: towards cellular native machine learning model splitting[J]. IEEE journal on selected areas in communications, 2022, 40(2): 626-640
[19] 孟祥福, 崔江燕, 邓敏超. 基于图卷积神经网络的最短路径距离估计方法[J]. 智能系统学报, 2024, 19(6): 1518-1527 MENG Xiangfu, CUI Jiangyan, DENG Minchao. Road network shortest distance estimation method based on graph convolutional networks[J]. CAAI transactions on intelligent systems, 2024, 19(6): 1518-1527
[20] 沈明辉, 刘宇杰, 陈婧, 等. 基于改进YOLOv8s轻量化网络的PCBA缺陷检测算法[J]. 计算机工程, 2026, 52(6): 326-338 SHEN Minghui, LIU Yujie, CHEN Jing, et al. Defect detection algorithm for PCBA based on improved YOLOv8s lightweight network[J]. Computer engineering, 2026, 52(6): 326-338
[21] 李沂杨, 陆声链, 王继杰, 等. 基于Transformer的DETR目标检测算法综述[J]. 计算机工程, 2026, 52(4): 62-81 LI Yiyang, LU Shenglian, WANG Jijie, et al. Review of DETR object detection algorithm based on transformer[J]. Computer engineering, 2026, 52(4): 62-81
[22] 李子奇, 苏宇轩, 孙俊, 等. 基于深度学习的多聚焦图像融合方法前沿进展[J]. 计算机科学与探索, 2024, 18(9): 2276-2292 LI Ziqi, SU Yuxuan, SUN Jun, et al. Critical review of multi-focus image fusion based on deep learning method[J]. Journal of frontiers of computer science & technology, 2024, 18(9): 2276-2292
[23] 吴一全, 蔡佳琦. 自动驾驶中深度学习的三维目标检测方法综述[J]. 智能系统学报, 2026, 21(2): 297-320 WU Yiquan, CAI Jiaqi. Deep learning-based 3D object detection for autonomous driving: a comprehensive review[J]. CAAI transactions on intelligent systems, 2026, 21(2): 297-320
[24] 田春伟, 宋明键, 左旺孟, 等. 卷积神经网络在图像超分辨上的应用[J]. 智能系统学报, 2025, 20(3): 719-749 TIAN Chunwei, SONG Mingjian, ZUO Wangmeng, et al. Application of convolutional neural networks in image super-resolution[J]. CAAI transactions on intelligent systems, 2025, 20(3): 719-749
[25] 张浩晢, 杨智博, 焦绪国, 等. 基于自优化神经网络的船舶运动模型辨识[J]. 智能系统学报, 2025, 20(3): 571-583 ZHANG Haozhe, YANG Zhibo, JIAO Xuguo, et al. Identification of ship motion model based on self-optimizing neural network[J]. CAAI transactions on intelligent systems, 2025, 20(3): 571-583
[26] 袁英华, 金英然, 高赟. 基于双维度特征增强的Transformer跟踪器[J]. 计算机工程, 2025, 51(12): 96-108 YUAN Yinghua, JIN Yingran, GAO Yun. Transformer tracker based on dual-dimensional feature enhancement[J]. Computer engineering, 2025, 51(12): 96-108
[27] 毕阳阳, 郑远帆, 史彩娟, 等. 基于深度学习的图像全景分割综述[J]. 计算机科学与探索, 2023, 17(11): 2605-2619 BI Yangyang, ZHENG Yuanfan, SHI Caijuan, et al. Survey on image panoptic segmentation based on deep learning[J]. Journal of frontiers of computer science and technology, 2023, 17(11): 2605-2619
[28] XU Xiaolong, TANG Sizhe, QI Lianyong, et al. CNN partitioning and offloading for vehicular edge networks in Web3[J]. IEEE communications magazine, 2023, 61(8): 36-42
[29] XIAO Yang, SONG Yuqian, LIU Jun. Collaborative multi-agent deep reinforcement learning for energy-efficient resource allocation in heterogeneous mobile edge computing networks[J]. IEEE transactions on wireless communications, 2024, 23(6): 6653-6668
[30] 车飞虎, 张大伟, 邵朋朋, 等. 基于四元数门控图神经网络的脚本事件预测[J]. 智能系统学报, 2023, 18(1): 138-143 CHE Feihu, ZHANG Dawei, SHAO Pengpeng, et al. Script event prediction based on a quaternion-gated graph neural network[J]. CAAI transactions on intelligent systems, 2023, 18(1): 138-143
Similar References:

Memo

-

Last Update: 2026-09-05

Copyright © CAAI Transactions on Intelligent Systems