IEEE TVT
2024
Crowdsourcing Live High Definition Map via Collaborative Computation in Automotive Edge Computing
Y. Zhang, Q. Liu, H. Wang, D. Chen, and K. Han
IEEE Transactions on Vehicular Technology, vol. 73, no. 9, pp. 13569–13583
Vehicular Edge Computing & Networking
Automated driving produces large, time-sensitive data flows, such as maps, shared perception, and model inference, that exceed what one vehicle can compute and what networks can deliver on demand. This work asks where computation should run, which data to send, and how to keep shared information fresh. It addresses these questions with methods that divide computation and data between vehicles and edge servers under latency and bandwidth limits.
Main themes
-
Live HD maps from vehicle and edge collaboration
Crowdsourced live high-definition maps built by collaborative computation between vehicles and edge servers, formulated to minimize transmission and computing resource usage under end-to-end latency limits; and COMAP, proactive provisioning of crowdsourced maps.
-
Real-time cooperative perception at the edge
AdaMap adaptively selects and compresses shared point-cloud data so that cooperative perception meets latency targets for large numbers of vehicles under changing network conditions.
-
Learning-based data delivery and offloading
DSORL, reinforcement-learning data-source selection for HD-map delivery in vehicular named data networking; REAL, learning-based offloading for edge-assisted mobile augmented reality; and network-aware task distribution for edge-assisted generative AI.
DSORL, IEEE T-ITS 2023 · REAL, IEEE TMC 2026 · Yan et al., IEEE ICC 2026
-
Vehicles as edge resources and information freshness
CAVE, which crowdsources passing-by vehicles for reliable in-vehicle edge computing, and age-of-information-aware service aggregation for connected vehicles.
Systems and validation
- Evaluation with an end-to-end 5G network simulator, a 5G emulation platform driven by a 150-vehicle CARLA trace, and a CARLA/ROS 2 vehicular named-data-networking environment. (Zhang et al., IEEE TVT 2024; AdaMap, ACM/IEEE SEC 2023; DSORL, IEEE T-ITS 2023)
Representative publications
A curated selection. Citations above link to each paper; see the full publication list for all work.
ACM/IEEE SEC
2023
AdaMap: High-scalable real-time cooperative perception at the edge
Q. Liu, Y. Xue, Y. Zhang, D. Chen, and K. Han
ACM/IEEE Symposium on Edge Computing (SEC), Wilmington, DE, USA
IEEE T-ITS
2023
DSORL: Data Source Optimization with Reinforcement Learning Scheme for Vehicular Named Data Networks
D. Doe, D. Chen, K. Han, H. Wang, J. Xie, and Z. Han
IEEE Transactions on Intelligent Transportation Systems, vol. 24, no. 10, pp. 11225–11237
IEEE TMC
2026Accepted
REAL: A Reinforcement Learning-based Intelligent Offloading Framework for Edge-Assisted MAR
L. Xie, A. Mallik, D. Chen, K. Han, and Z. Han
IEEE Transactions on Mobile Computing
IEEE GLOBECOM
2024
CAVE: Crowdsourcing passing-by vehicles for reliable in-vehicle edge computing
J. Cao, Q. Liu, D. Chen, and K. Han
IEEE Global Communications Conference: Mobile and Wireless Networks (GLOBECOM), Cape Town, South Africa
IEEE ICC
2026
Network-Aware Intelligent Task Distributor for Edge-Assisted Generative AI
J. Yan, M. Talukder, D. Chen, K. Han, and L. Xie
IEEE International Conference on Communications (ICC), Glasgow, United Kingdom