Distributed Dynamic Map Fusion via Federated Learning for Intelligent Networked Vehicles

Zhang, Zijian; Wang, Shuai; Hong, Yuncong; Zhou, Liangkai; Hao, Qi

doi:10.1109/icra48506.2021.9561612

Cited by 43 publications

(13 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For future work, we will first reduce the size of SCST for the larger scale of the edge server. Next, machine learning and federated learning can be introduced to improve the performance of edge servers, good solutions can be found in [ 50 , 51 , 52 ]. In addition, different regions have different traffic rules and habits, these should be considered.…”

Section: Discussionmentioning

confidence: 99%

RPVC: A Revocable Publicly Verifiable Computation Solution for Edge Computing

Jiao

Zhou

Wang

et al. 2022

Sensors

View full text Add to dashboard Cite

With publicly verifiable computation (PVC) development, users with limited resources prefer to outsource computing tasks to cloud servers. However, existing PVC schemes are mainly proposed for cloud computing scenarios, which brings bandwidth consumption or network delay of IoT devices in edge computing. In addition, dishonest edge servers may reduce resource utilization by returning unreliable results. Therefore, we propose a revocable publicly verifiable computation(RPVC) scheme for edge computing. On the one hand, RPVC ensures that users can verify the correct results at a small cost. On the other hand, it can revoke the computing abilities of dishonest edge servers. First, polynomial commitments are employed to reduce proofs’ length and generation speed. Then, we improve revocable group signature by knowledge signatures and subset covering theory. This makes it possible to revoke dishonest edge servers. Finally, theoretical analysis proves that RPVC has correctness and security, and experiments evaluate the efficiency of RPVC.

show abstract

Section: Discussionmentioning

confidence: 99%

RPVC: A Revocable Publicly Verifiable Computation Solution for Edge Computing

Jiao

Zhou

Wang

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Notice that the raw data generated from CARLA are not directly compatible with SECOND. To address this issue, we develop a python-based data transformation module, such that the transformed dataset meets the KITTI standard [45]- [47]. The federated learning model training is implemented using PyTorch with python 3.8 on Linux server with an NVIDIA RTX 3090 GPU.…”

Section: Tofel For Autonomous Drivingmentioning

confidence: 99%

Accelerating Federated Edge Learning via Topology Optimization

Huang¹,

Zhang²,

Wang³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Federated edge learning (FEEL) is envisioned as a promising paradigm to achieve privacy-preserving distributed learning. However, it consumes excessive learning time due to the existence of straggler devices. In this paper, a novel topology-optimized federated edge learning (TOFEL) scheme is proposed to tackle the heterogeneity issue in federated learning and to improve the communication-and-computation efficiency. Specifically, a problem of jointly optimizing the aggregation topology and computing speed is formulated to minimize the weighted summation of energy consumption and latency. To solve the mixed-integer nonlinear problem, we propose a novel solution method of penalty-based successive convex approximation, which converges to a stationary point of the primal problem under mild conditions. To facilitate real-time decision making, an imitationlearning based method is developed, where deep neural networks (DNNs) are trained offline to mimic the penalty-based method, and the trained imitation DNNs are deployed at the edge devices for online inference. Thereby, an efficient imitate-learning based approach is seamlessly integrated into the TOFEL framework. Simulation results demonstrate that the proposed TOFEL scheme accelerates the federated learning process, and achieves a higher energy efficiency. Moreover, we apply the scheme to 3D object detection with multi-vehicle point cloud datasets in the CARLA simulator. The results confirm the superior learning performance of the TOFEL scheme over conventional designs with the same resource and deadline constraints.

show abstract

“…Peng et al [45] introduced an adaptive FL framework for autonomous vehicles. In [46], the authors addressed the problem of distributed dynamic map fusion with FL for intelligent networked vehicles.…”

Section: Related Workmentioning

confidence: 99%

Deep Federated Learning for Autonomous Driving

Nguyen¹,

Do²,

Tran³

et al. 2021

Preprint

View full text Add to dashboard Cite

Autonomous driving is an active research topic in both academia and industry. However, most of the existing solutions focus on improving the accuracy by training learnable models with centralized large-scale data. Therefore, these methods do not take into account the user's privacy. In this paper, we present a new approach to learn autonomous driving policy while respecting privacy concerns. We propose a peer-to-peer Deep Federated Learning (DFL) approach to train deep architectures in a fully decentralized manner and remove the need for central orchestration. We design a new Federated Autonomous Driving network (FADNet) that can improve the model stability, ensure convergence, and handle imbalanced data distribution problems while is being trained with federated learning methods. Intensively experimental results on three datasets show that our approach with FADNet and DFL achieves superior accuracy compared with other recent methods. Furthermore, our approach can maintain privacy by not collecting user data to a central server. Our source code and trained models are available at https://github.com/aioz-ai/FADNet

show abstract

Distributed Dynamic Map Fusion via Federated Learning for Intelligent Networked Vehicles

Cited by 43 publications

References 22 publications

RPVC: A Revocable Publicly Verifiable Computation Solution for Edge Computing

RPVC: A Revocable Publicly Verifiable Computation Solution for Edge Computing

Accelerating Federated Edge Learning via Topology Optimization

Deep Federated Learning for Autonomous Driving

Contact Info

Product

Resources

About