Onboard Data Store Management for Autonomous Vehicle Capacity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles generate vast amounts of data during operation, which are typically stored onboard until the vehicle returns to a facility for retrieval and analysis, necessitating a robust and efficient system for secure and real-time data management to ensure continuous storage capacity and facilitate immediate data access.
Innovation Solution
A framework and system that automatically manages data storage by monitoring available capacity and annotating data for deletion based on age and priority, ensuring sufficient free space and allowing for real-time data transmission to remote monitoring systems, thereby maintaining efficient data storage and enabling immediate access to relevant data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is stored onboard the autonomous vehicle during operational runs, then data security and continuous storage capacity are ensured, but data retrieval requires vehicle return to facility and physical removal of storage devices
Solution Approach 1:
The patent extracts and transmits only the essential data elements (annotations and associated data) from the vehicle's onboard storage to the remote server, separating the critical information retrieval function from the physical vehicle return requirement. This allows remote access to annotated data without needing to physically remove storage devices or wait for vehicle return.
Solution Approach 2:
The patent introduces a remote server as an intermediary between the vehicle's onboard storage and the data analysis system. The server receives transmissions from the vehicle, stores annotated data, and provides remote access capabilities, acting as a mediator that eliminates the need for direct physical connection or vehicle return for data retrieval.
2Quantity of substance
If vast amounts of sensor data are continuously stored onboard the autonomous vehicle, then complete operational data is preserved, but storage capacity may be exhausted requiring data deletion
Solution Approach 1:
The patent implements preliminary annotation of data during vehicle operation, identifying and marking important data elements before storage capacity is exhausted. This preliminary identification allows prioritized transmission and retention of critical data, ensuring that when deletion becomes necessary, only non-essential data is removed while preserving annotated important data.
Solution Approach 2:
The patent applies different quality characteristics to different data elements through annotation, where important data is marked with specific metadata indicating its significance. This allows the system to treat annotated data differently from unannotated data, preserving high-quality annotated data while allowing deletion of lower-priority unannotated data when storage capacity is limited.
3Loss of information
If engineers wait for vehicle return to facility to retrieve and analyze data, then complete operational data is available for analysis, but operational efficiency is reduced due to delayed data access
Solution Approach 1:
The patent extracts and transmits annotated data elements to a remote server during vehicle operation or at convenient intervals, allowing engineers to access and analyze this subset of data remotely without waiting for vehicle return. This extraction approach enables continuous analysis of important data while the vehicle remains in operation.
Solution Approach 2:
The patent establishes a feedback loop where annotated data is transmitted to the remote server, analyzed by engineers, and used to generate insights that can inform future vehicle operations. This continuous feedback mechanism improves productivity by enabling real-time or near-real-time analysis rather than batch processing after vehicle return.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
A system and method including monitoring an available storage capacity of a data store onboard a vehicle and annotated data stored therein, the annotated data stored in the data store including event data associated with an operation of the vehicle; automatically deleting, in response to an indication of at least one of the available storage capacity of the data store and the annotated data stored therein, at least a portion of the data from the data store; and generating an output including an indication of a completion of the automatic deleting of the at least a portion of the data.