.png){kind=icon}
.png){kind=icon}
Return
A survey on dependent task offloading in vehicular edge computing: models, algorithms, and challenges
.png){kind=icon}
PRE
AI
C
Z
X
DOI：10.1007/s10586-026-06451-9.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Dependent task offloading (DTO) is a critical problem in vehicular edge computing (VEC), yet it faces a fundamental triad of constraints to balance complex task dependencies, highly dynamic environmental changes, and stringent real-time latency requirements. Despite the advancements in the field, previous surveys either focus on general edge computing or lack a systematic analysis of how different algorithms manage trade-offs among these interconnected constraints. To address this gap, wo aim to provide a systematic review of DTO in vehicular environments and proposes a synthesized five-dimensional dependency taxonomy specifically designed for VEC. Based on this taxonomy, a unified analytical perspective is developed. It incorporates the Directed Acyclic Graph (DAG) model and its extensions, including time-varying attributes and probabilistic dependencies. This enables more accurate modeling of dynamic task relationships. Using this unified analytical foundation, DTO is examined from a process-oriented perspective, covering dependency modeling, scheduling optimization, resource allocation, and dynamic adaptation. We systematically compare mainstream algorithmic paradigms to illustrate how their underlying mechanisms can result in distinct trade-off strategies and applicability boundaries under the dynamics–dependency–real-time triad of constraints. Finally, key limitations in current research are characterized and future directions for building dependency-aware, adaptive, and scalable vehicular offloading systems are outlined. Overall, this survey aims to offer a unified perspective for understanding DTO and provide methodological guidance for designing next-generation VEC systems.
Keywords:
Vehicular Edge Computing
Task Dependency
Task Offloading
Dependency Modeling
Algorithm
Journal
C
IF:
4.1
Papers:
4.8K
Citations:
7.5K
Organization
Cited Papers