Vehicle Control Model Optimization via Data Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle control systems require large amounts of data and processing time for optimizing steering and road information, and expose personal user information when accessing driving data.
Innovation Solution
A method and apparatus that involve obtaining vehicle control parameters, transmitting them to a server for optimization, training a model based on user driving information, and controlling the vehicle using the optimized model, while protecting personal information by processing and encrypting data before transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If large amounts of driving data are collected and processed locally to optimize vehicle control, then control accuracy is improved, but processing time increases and personal information privacy is exposed
Solution Approach 1:
The patent introduces an intermediate processing layer that extracts and transmits only essential driving features and parameters to the server, rather than transmitting raw driving data. This intermediary processing step reduces the data volume significantly while preserving the information needed for accurate vehicle control optimization.
Solution Approach 2:
The system extracts only the necessary driving features and intermediate output values from the complete driving data set before transmission. By taking out only the essential information elements required for model training and control optimization, the system reduces processing time and data transmission requirements while maintaining control accuracy.
2Measurement precision
If complete driving data is transmitted to server for model optimization, then control accuracy is improved, but personal information privacy is exposed
Solution Approach 1:
The system extracts and transmits only anonymized driving features and intermediate output values, removing all personally identifiable information from the transmitted data. This extraction process ensures that only the necessary technical parameters for model optimization are shared, protecting user privacy while maintaining control accuracy.
Solution Approach 2:
An intermediate processing layer acts as a mediator between the local driving data and the remote server, transforming raw driving data into anonymized feature representations. This intermediary transformation protects personal information by ensuring that only processed, anonymized data leaves the local system.
3Speed
If raw driving data is processed locally without optimization, then processing speed is maintained, but data volume requirements increase
Solution Approach 1:
The system extracts only the essential driving features and intermediate output values from the complete driving data set, removing redundant and unnecessary information. This extraction reduces the data volume significantly while maintaining the information needed for effective model training and control optimization.
Solution Approach 2:
The driving data processing is segmented into multiple stages: local feature extraction, intermediate output generation, selective transmission to server, and remote model optimization. This segmentation allows speed-critical operations to occur locally while data-intensive operations occur remotely, optimizing both processing speed and data volume management.
Data Source
AI summary
A method of controlling a vehicle using a first model trained according to optimized output of a second model.