Distributed Vehicle Data Processing via Edge Server Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data transfer mechanisms for autonomous vehicles, such as physical transfer of hard disks or network uploads, are limited by network bandwidth and computational capabilities, preventing real-time analysis and processing of vehicle data.
Innovation Solution
A distributed computing system that determines resource requirements for vehicle data and selects an optimal computation resource based on selection policies, such as computational and storage needs, proximity, and geographic boundaries, to efficiently process and utilize vehicle data for applications like autonomous navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data is transferred using physical hard disk transfer or network uploads, then data can be moved from vehicle to processing resource, but network bandwidth and computational capabilities are limited preventing real-time analysis
Solution Approach 1:
The system segments the data processing function across multiple distributed computation resources (edge servers and cloud servers) rather than relying on a single centralized resource. This allows the vehicle to select and distribute data processing tasks across multiple servers simultaneously, improving overall processing throughput and reducing latency through parallel processing capabilities
Solution Approach 2:
The system introduces an intermediary selection mechanism that intelligently routes data to appropriate computation resources based on real-time conditions. The vehicle controller acts as an intermediary that evaluates available computation resources and selects the optimal server for processing specific data types, enabling efficient real-time analysis while managing network bandwidth constraints
2Ease of operation
If data is processed offline on servers, then computational load on vehicle is reduced, but real-time analysis and processing capabilities are lost
Solution Approach 1:
The system implements local quality by deploying computation resources at multiple locations with different capabilities - edge servers positioned near vehicles provide low-latency processing for time-critical data, while cloud servers handle less time-sensitive tasks. This spatial differentiation of computation quality allows the vehicle to maintain low computational load while achieving real-time processing when needed by selecting appropriate edge resources
3Productivity
If computation resources are distributed across multiple locations, then processing capacity increases, but selecting the optimal resource becomes more complex
Solution Approach 1:
The system manages selection complexity by dynamically changing evaluation parameters based on data type and urgency. The vehicle controller adjusts selection criteria such as network proximity, server computational capacity, and current load conditions depending on the specific processing requirements. This parameter adaptation simplifies the selection process for each task while still utilizing the full distributed processing capacity
Data Source
AI summary
The technology disclosed allows for remote distributed computing and storage of vehicle data obtained from one or more vehicles. Specifically, the technology disclosed is capable of determining, at a vehicle, resource requirements for a software application. The technology disclosed is also capable of selecting a computation resource from a plurality of computation resources based on one or more selection policies for meeting the resource requirements, sending the vehicle data from the vehicle to the selected computation resource, receiving at the vehicle data from the selected computation resource and utilizing the data obtained from the computation resource and data in the vehicle in the software application.