Vehicle Data Set Evaluation for Driver Assistance Storage Efficiency

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing vehicle systems face challenges in efficiently handling and storing large amounts of data, struggling to identify and utilize relevant data for assistive and autonomous driving, leading to increased storage costs and inefficiencies.

Innovation Solution

A method for evaluating a trained vehicle data set by receiving sensor data, running scene detection operations, scoring runtime attributes against a vector representation, and determining event flags to identify relevant data and update the data set, thereby improving data handling and storage efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If existing systems collect and store large amounts of data indiscriminately, then the data storage capacity increases, but the cost for storing data and the difficulty in identifying relevant data increase significantly

Engineering Contradiction:

Improvedata storage capacityVSAvoiddifficulty in identifying relevant data

Core Design Contradiction:

Quantity of substanceVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by evaluating the trained data set before full deployment, scoring run-time attributes against a vector representation to identify which data will be relevant. This allows the system to pre-determine data importance metrics and event flags, avoiding the need to process and identify relevant data from indiscriminately collected data during operation.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system extracts only the essential and relevant features from the collected data by scoring run-time attributes against a vector representation. This extraction process identifies target object attributes and generates event flags that highlight only the significant data points, separating relevant information from the overwhelming volume of collected data.

Inventive Principle:

Principle #2Taking out (Extraction)

2Quantity of substance

If existing systems collect and store large amounts of data indiscriminately, then the data storage capacity increases, but the storage cost increases significantly

Engineering Contradiction:

Improvedata storage capacityVSAvoidstorage cost

Core Design Contradiction:

Quantity of substanceVSLoss of energy

Solution Approach 1:

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system discards irrelevant data by not storing it in the first place, using the evaluation framework to determine which data points are worth keeping. The event flags identify only the data samples that need to be retained for updating the trained data set, effectively discarding the majority of indiscriminately collected data that would otherwise consume storage resources.

Inventive Principle:

Principle #34Discarding and recovering

3Quantity of substance

If existing systems generate and store data indiscriminately, then the data volume increases, but the ability to handle data efficiently decreases

Engineering Contradiction:

Improvedata volumeVSAvoiddata handling efficiency

Core Design Contradiction:

Quantity of substanceVSProductivity

Solution Approach 1:

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Inventive Principle:

Principle #2Taking out (Extraction)

4Device complexity

If existing systems do not provide configurations to interpret significant data, then the system simplicity is maintained, but the ability to identify relevant data decreases

Engineering Contradiction:

Improvesystem configuration complexityVSAvoidability to identify relevant data

Core Design Contradiction:

Device complexityVSLoss of information

Solution Approach 1:

The system implements feedback mechanisms where the evaluation of run-time attributes against the vector representation provides information about data significance. The scoring process generates feedback in the form of event flags that indicate which data points are relevant, allowing the system to continuously improve its ability to identify significant data without requiring complex manual configurations.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS10839263B2System and method for evaluating a trained vehicle data set familiarity of a driver assitance system

Publication Date: 2020.11.17 HARMAN INT IND INC

AI summary

The present disclosure relates to systems, devices and methods for evaluating a trained vehicle data set of a driver assistance system. Embodiments are directed to scoring run time attributes of a scene detection operation using a trained vehicle data set against a vector representation for an annotated data set to assess the ability of the scene detection operation to perceive target object attributes of the vehicle sensor data. In one embodiment, scoring evaluates effectiveness of the scene detection operation in identifying target object attributes of the vehicle sensor data using the trained vehicle data set. An event flag may be determined for a trained vehicle data set based on the scoring, the even flag identifying one or more parameters for updating the trained vehicle data set. Configurations and processes can identify anomalies to a trained vehicle data set and allow for capturing useful real world data to test runtime operations.