In-Vehicle Control Model Updates for Real-Time ECU Adaptation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing electronic control systems in vehicles face challenges in adapting to real-time changes in vehicle behavior due to the limitations of incremental software updates, which fail to effectively leverage the vast data exchanged between vehicle functional units.

Innovation Solution

The implementation of distributed and federated machine intelligence systems that utilize a model manager, control component, and learning component to construct and modify trainable models based on observational data, allowing for dynamic calibration of vehicle functional units and adaptation to changing conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If incremental software updates are used to maintain vehicle operation, then system reliability is improved, but the system cannot adapt in real-time to changing vehicle behavior

Engineering Contradiction:
Improvevehicle operation safetyVSAvoidreal-time adaptation capability
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent transforms the static incremental update approach into a dynamic system where machine learning models continuously learn from observational data exchanged between vehicle functional units. The system adapts in real-time by updating models based on changing vehicle behavior patterns, while maintaining reliability through the structured update framework.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system implements feedback mechanisms where observational data from vehicle functional units is collected, processed through machine learning models, and used to generate updated control parameters. This closed-loop feedback enables continuous adaptation while maintaining system reliability through validated update processes.

Inventive Principle:
Principle #23Feedback

2Adaptability or versatility

If machine learning models are trained using vast amounts of exchanged data, then adaptability is improved, but data processing complexity increases

Engineering Contradiction:
Improvereal-time adaptation capabilityVSAvoiddata processing complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the data processing task by distributing machine learning model training across multiple vehicle functional units. Each unit contributes observational data and computes local model updates, reducing the processing burden on any single component while collectively achieving comprehensive adaptability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system creates a universal machine learning framework that can process diverse observational data from various vehicle functional units through a common model structure. This multi-functional approach handles different data types and sources uniformly, reducing overall system complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP4092577A1Real-time in-vehicle modeling and simulation updates
Publication Date: 2022.11.23 VOLVO CAR CORP
  • EP4092577A1 patent drawingFigure 1
  • EP4092577A1 patent drawingFigure 2
  • EP4092577A1 patent drawingFigure 3~4

AI summary

Systems, devices, computer-implemented methods, and/or computer program products that facilitate modifying electronic control system behavior using distributed and/or federated machine intelligence. In one example, a system can comprise a process that executes computer executable components stored in memory. The computer executable components can comprise a model manager, a control component, and a learning component. The model manager can construct a trainable model using a pre-trained template model that is received via an in-vehicle network from a domain chief. The control component can dynamically vary a control parameter of a vehicle functional unit using the pre-trained template model to calibrate an output of the vehicle functional unit. The learning component can modify the trainable model based on observational data of the vehicle functional unit.