Autonomous Vehicle Data Ingestion Platform
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Developing artificial intelligence (AI) and machine learning (ML) models for autonomous vehicles is a time-consuming and complex process, often requiring manual and disparate tools, leading to errors and increased development time due to the lack of an integrated platform for data ingestion, processing, and model development.
Innovation Solution
A data science system that provides an end-to-end platform for ingesting, processing, and visualizing data, automating tasks, and provisioning resources, allowing for seamless transition through the development cycle and integration of various tools and services within a single ecosystem.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual processes are used for data collection and model development, then flexibility and customization are improved, but development time and error rates increase
Solution Approach 1:
The system enables self-service through automated data collection from vehicles, automatic data cleaning and processing, and self-organizing of training datasets. The platform autonomously performs tasks that would otherwise require manual intervention by data scientists, thereby reducing development time while maintaining operational flexibility through configurable parameters and options.
2Adaptability or versatility
If multiple disparate tools are used for data processing and model development, then specific task capabilities are improved, but system complexity and integration difficulty increase
Solution Approach 1:
The patent merges multiple disparate tools and functions into a single integrated platform. The system combines data collection from vehicles, data cleaning, data processing, model training, and evaluation into one unified ecosystem, eliminating the need to switch between multiple separate tools while maintaining all necessary task capabilities.
Solution Approach 2:
The platform achieves universality by designing a multi-functional system that can perform diverse data science tasks through a common architecture. The system handles various data types, supports multiple model training approaches, and provides versatile data processing capabilities all within a single unified platform, reducing system complexity while maintaining adaptability.
3Reliability
If comprehensive data collection from vehicles is performed, then model training quality is improved, but data processing complexity and resource requirements increase
Solution Approach 1:
The system performs preliminary actions by automatically cleaning, validating, and organizing data as it is collected from vehicles. Data processing operations are initiated and completed in advance before model training begins, ensuring high-quality training datasets are ready without adding complexity to the training process itself. This preliminary data preparation maintains model quality while simplifying the overall workflow.
Data Source
AI summary
In one embodiment, a method includes obtaining candidate data generated by a vehicle. The candidate data comprises a subset of sensor data identified based on a set of neural network models executing on the vehicle. The method also includes determining whether the candidate data can be associated with one or more categories of a set of categories for training data based on a set of categorization models. The method further includes associating the candidate data with the first category in response to determining that the candidate data can be associated with at a first category of the set of categories. The method further includes determining whether the candidate data can be associated with a second category. The set of categories lacks the second category. The method further includes including the second category in the set of categories in response to determining that the candidate data can be associated with the second category. The method further includes associating the candidate data with the second category.