Vehicular Fog Task Scheduling for Low-Latency ADAS Offloading
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems lack an efficient method to schedule the offloading of complex computing and control tasks from vehicles to fog servers in vehicular networks, particularly in environments with dense traffic flow.
Innovation Solution
A system and method that utilize cellular vehicle-to-everything (C-V2X) communication and fog nodes (RSUs) to schedule offloading of vehicle tasks, employing a Hungarian-based algorithm to minimize total task delay by dynamically maintaining records of task data rates and delay times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If cloud computing is used to perform computationally intensive tasks, then processing power is improved, but network latency and round-trip delays increase
Solution Approach 1:
The patent introduces fog nodes as intermediary computing resources positioned between cloud data centers and vehicle edge devices. These fog nodes provide distributed computing power closer to the vehicles, enabling computationally intensive ADAS tasks to be processed with reduced network latency compared to distant cloud servers, while maintaining the processing capabilities that vehicles lack
Solution Approach 2:
The patent adds a spatial dimension to the computing architecture by deploying fog nodes at multiple geographic locations along road networks. This transforms the single-point cloud computing model into a distributed multi-dimensional computing system, allowing vehicles to access computing resources at nearby fog nodes rather than relying on distant cloud data centers
2Productivity
If more fog nodes are deployed to handle dense traffic flow, then task processing capacity is improved, but system complexity increases
Solution Approach 1:
The patent segments the centralized cloud computing function into multiple distributed fog nodes deployed along the road network. Each fog node independently handles computing tasks for nearby vehicles, dividing the overall processing capacity across multiple smaller units. This segmentation increases total task processing capacity while managing complexity through modular, distributed architecture rather than a single complex centralized system
3Loss of time
If vehicles offload tasks to nearby fog nodes, then network delay is reduced, but task scheduling complexity increases
Solution Approach 1:
The patent implements preliminary action through the Hungarian algorithm, which pre-calculates and determines the optimal assignment of vehicles to fog nodes based on current network conditions, vehicle locations, and task characteristics. By performing this optimization in advance rather than reactively, the system reduces scheduling complexity while maintaining low network delays for task offloading
Data Source
AI summary
A system for scheduling vehicle tasks for vehicles is described. The system includes a plurality of vehicles having the advanced driver assistance systems and cellular vehicle-to-everything communication, and a plurality of road side unit (RSU) nodes arranged at predetermined locations to execute driver assist tasks. The system further includes a central base station configured to receive and dynamically maintain a record of task data rates between the vehicles and the RSUs and delay time to return an executed task, and schedule offloading of the tasks for the advanced driver assistance systems. The RSUs are fog nodes for computing the tasks in accordance with the schedule, and the tasks are control tasks for the advanced driver assistance systems.