Edge Computing Ground Truth Service for ADAS Validation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Automated driver-assistance systems (ADAS) face challenges in accurately modeling dynamic and static objects in a local environment due to the lack of reliable ground truth information, which hinders continuous validation and learning in changing conditions, potentially compromising vehicle safety.

Innovation Solution

A network-based ground truth information service utilizing edge computing devices aggregates and verifies high-confidence data from various sources, including vehicles and roadside sensors, to create datasets that can be distributed to supervised learning systems for validation and improvement.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If supervised machine learning techniques are used to generate environment models in real-time, then the system can adapt to dynamically changing conditions, but the system lacks reliable ground truth information for validation, compromising safety

Engineering Contradiction:

Improveadaptability to dynamically changing conditionsVSAvoidreliability of environment model

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system performs preliminary actions by collecting and storing ground truth information from trusted sources (GPS data, sensor measurements) before validation is needed. This pre-collected ground truth data is then available for validating the environment model when adaptation to changing conditions occurs, ensuring both adaptability and reliability

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback by using ground truth information to validate environment model outputs and update the model accordingly. This closed-loop validation process ensures that the adaptive system maintains reliability by continuously comparing its perceptions against known ground truth data and learning from discrepancies

Inventive Principle:

Principle #23Feedback

2Measurement precision

If ground truth information is collected from multiple sources to improve validation accuracy, then measurement precision improves, but device complexity and data processing requirements increase

Engineering Contradiction:

Improveprecision of ground truth informationVSAvoidcomplexity of data aggregation system

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system applies universality by using a standardized data collection and validation framework that can work with multiple different data sources (GPS, sensors, maps) through common interfaces and protocols. This multi-functional approach enables precise ground truth validation without proportionally increasing system complexity, as the same validation mechanisms handle diverse data types

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS12082082B2Validation and training service for dynamic environment perception based on local high confidence information

Publication Date: 2024.09.03 INTEL CORP

AI summary

Described herein is a high confidence ground truth information service executing on a network of edge computing devices. A variety of participating devices obtain high confidence ground truth information relating to objects in a local environment. This information is communicated to the ground truth information service, where it may be verified and aggregated with similar information before being communicated as part of an acquired ground truth dataset to one or more subscribing devices. The subscribing devices use the ground truth information, as included in the ground truth dataset, to both validate and improve their supervised learning systems.