Edge Data Offloading for Autonomous Mobility Resource Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computing and data storage architectures are inefficient for autonomous mobility implementations, such as autonomous vehicles, which generate large amounts of data, as they struggle with onsite storage and transmitting data to remote cloud systems due to bandwidth constraints and high computational costs.
Innovation Solution
Implementing a local data acquisition and computing system with content absorption nodes near the data source, allowing for rapid data offloading and real-time processing, reducing the need for remote data transmission and leveraging regional and global scheduling layers for efficient data management and processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data is transmitted from autonomous mobility edge devices to remote cloud storage systems, then data can be stored and processed, but bandwidth constraints severely limit the transmission capability
Solution Approach 1:
The patent segments the centralized cloud storage system into distributed edge storage nodes located near autonomous vehicles. These edge nodes cache frequently accessed data locally, reducing the volume of data that needs to be transmitted over bandwidth-constrained networks while maintaining access to large datasets for model training and inference.
Solution Approach 2:
The patent implements local quality by deploying storage and computing resources at the edge of the network close to data sources. This allows autonomous vehicles to access frequently used data locally without requiring high-bandwidth remote connections, while still leveraging centralized cloud resources when needed.
2Productivity
If applications are deployed on autonomous vehicles to process collected data, then real-time processing is enabled, but the data rate increase outstrips any infrastructure investment considered feasible
Solution Approach 1:
The patent extracts heavy data processing and model training workloads from autonomous vehicles and relocates them to centralized cloud computing infrastructure. This allows vehicles to maintain simpler onboard systems for real-time inference while leveraging powerful remote resources for computationally intensive tasks like model retraining and large-scale data analysis.
Solution Approach 2:
The patent introduces a hierarchical architecture with multiple dimensions of computing power distribution. Instead of a single dimension of either all processing at the vehicle or all processing in the cloud, the system creates a multi-dimensional framework where different types of processing occur at different levels based on their requirements for speed, power, and data access patterns.
3Ease of manufacture
If the vehicle is in proximity to the cloud infrastructure, then the deployment cost is lowest, but autonomous vehicles operate remotely from cloud systems
Solution Approach 1:
The patent introduces edge computing nodes and gateway systems as intermediaries between autonomous vehicles and centralized cloud infrastructure. These intermediaries provide local caching, preprocessing, and coordination capabilities that reduce both the operational latency for vehicles and the infrastructure investment required compared to purely remote cloud-based systems.
Data Source
AI summary
A system and method that enables resource analysis within an intelligent content absorption network with autonomous mobility implementations includes generating a prioritization model, training the prioritization model on a data corpus to generate a prioritization schedule, deploying the prioritization schedule to one or more autonomous vehicles, offloading data at a local content absorption node, processing data at the local content absorption node, and optionally transmitting data results from the processed data.