Vehicular Micro Cloud Task Assignment by Mobility
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
It is challenging to determine how to efficiently assign computational tasks to members of a vehicular micro cloud, particularly in predicting which members have the most computing resources available and assigning tasks based on these predictions.
Innovation Solution
The system determines vehicle travel speeds within a vehicular micro cloud and assigns computational sub-tasks to members that are stationary or traveling at the slowest speeds, using a leader vehicle to manage task distribution based on factors like processing power, sensor accuracy, and memory availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If computational tasks are assigned to moving vehicles in a vehicular micro cloud, then task distribution can be achieved, but computational efficiency and reliability deteriorate due to vehicle mobility and resource variability
Solution Approach 1:
The system performs preliminary actions by predicting future computing resource availability of vehicles before assigning computational tasks. The prediction module forecasts resource availability based on historical data and current vehicle states, allowing the system to proactively identify suitable candidate vehicles for task execution, thereby ensuring both efficiency and reliability
Solution Approach 2:
The system implements feedback mechanisms where actual task execution results and resource consumption data are collected and fed back to the prediction module. This feedback loop continuously refines the prediction accuracy of computing resource availability, enabling dynamic optimization of task assignment decisions to maintain high computational efficiency and reliability
2Reliability
If computational tasks are assigned to stationary or slow-moving vehicles, then task execution stability improves, but task distribution flexibility and speed worsen
Solution Approach 1:
The system applies dynamics by making the task assignment strategy adaptive rather than static. The prediction module continuously monitors vehicle mobility patterns and resource availability, dynamically adjusting task assignment decisions. This allows the system to flexibly select between stationary and moving vehicles based on real-time conditions, balancing execution stability with distribution speed
Solution Approach 2:
The system changes parameters by considering multiple vehicle attributes including mobility state, computing resource availability, and task complexity. By varying the weighting of these parameters based on current system needs, the system can optimize for either stability or speed depending on the situation, resolving the contradiction between execution stability and distribution efficiency
3Productivity
If task assignment is based on predicted computing resource availability, then resource utilization efficiency improves, but system complexity and prediction accuracy requirements worsen
Solution Approach 1:
The system segments the complex prediction problem into manageable components by evaluating individual vehicle attributes separately (mobility state, resource availability, historical performance) and combining them through weighted scoring. This segmentation approach simplifies the overall prediction system while maintaining high resource utilization efficiency
Solution Approach 2:
Vehicles autonomously provide information about their computing resource availability and mobility state to the prediction system. This self-service mechanism reduces the complexity of centralized monitoring by leveraging the vehicles' own capabilities to report their status, thereby simplifying the prediction system architecture while improving resource utilization
Data Source
AI summary
The disclosure includes embodiments for a set of connected vehicles to collectively execute tasks which no single vehicle can execute due to computational limitations of the single vehicle. In some embodiments, a method includes determining, for a vehicular micro cloud, a set of computing sub-tasks to be completed. The method includes determining vehicle travel speeds for the members of the vehicular micro cloud. The method includes assigning the computing sub-tasks to the members based on the vehicle travel speeds of the members relative to one another so that the members that the computational sub-tasks are assigned to the members that are either stationary or traveling at the slowest vehicle travel speeds. The computing sub-task is completed by the member to which it is assigned.