Autonomous Vehicle Data Orchestration for Selective Edge-to-Cloud Transfer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The management of vast amounts of data generated by autonomous vehicles poses challenges in terms of safety, security, cost-effectiveness, and scalability, particularly in ensuring timely and secure data transmission and storage across various priorities.
Innovation Solution
A data management system that includes a data orchestrator onboard the vehicle, capable of processing and transmitting vehicle data by generating metadata, using predictive models to determine which data to transmit to which entities and when, and employing edge intelligence for real-time data orchestration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If all vehicle data is transmitted to cloud data centers, then data availability for analysis and model training is improved, but transmission costs and network bandwidth requirements increase significantly
Solution Approach 1:
The patent extracts only the necessary subset of vehicle data for transmission to cloud data centers, rather than transmitting all generated data. Edge computing devices process and filter data locally, extracting only relevant information for cloud analysis, thereby reducing transmission costs while maintaining data availability for essential operations.
Solution Approach 2:
The patent segments data management into multiple levels: local vehicle processing, edge computing processing, and cloud data center processing. This segmentation allows different types of data to be handled at appropriate levels, reducing the burden on cloud transmission while ensuring critical data is properly analyzed and stored.
2Reliability
If a centralized cloud data center stores all vehicle data, then data security and centralized management are improved, but storage costs and data access latency increase
Solution Approach 1:
The patent implements a nested data storage architecture where vehicle data is stored at multiple hierarchical levels: local vehicle storage, edge computing storage, and cloud data center storage. This nested structure allows fast local access for time-critical operations while maintaining centralized secure storage in the cloud for long-term retention and security management.
Solution Approach 2:
The patent performs preliminary data processing and filtering at the edge computing level before data is archived in cloud data centers. This preliminary action reduces the volume of data requiring centralized storage while ensuring that security-critical data is properly prepared and validated before centralized archiving, reducing both storage costs and access latency.
3Speed
If edge computing devices process data in real-time, then data processing speed and responsiveness are improved, but device complexity and computational requirements increase
Solution Approach 1:
The patent implements partial processing at edge devices, where only critical real-time processing functions are executed locally while less time-sensitive processing is performed in the cloud. This partial action approach provides real-time responsiveness for safety-critical operations without requiring edge devices to have excessive computational complexity for all processing tasks.
Data Source
AI summary
The present disclosure provides methods and systems for managing autonomous vehicle data. The method may comprise: (a) collecting said autonomous vehicle data from the autonomous vehicle, wherein the autonomous vehicle data has a size of at least 1 terabyte; (b) processing the autonomous vehicle data to generate metadata corresponding to the autonomous vehicle data, wherein the autonomous vehicle data is stored in a database; (c) using at least a portion of the metadata to retrieve a subset of the autonomous vehicle data from the database, which subset of the autonomous vehicle data has a size less than the autonomous vehicle data; and (d) storing or transmitting the subset of the autonomous vehicle data.