Switchable Autonomous Driving Simulation Models for Lower Test Load

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current autonomous driving simulation systems are limited by their inability to integrate variably-sourced simulation data models, leading to unnecessary costs, processing burdens, and extended testing times due to the inclusion of non-essential models and high accuracy sensors for all applications.

Innovation Solution

A customizable autonomous driving simulation architecture that allows for the integration of multiple simulation data models from different sources, enabling the selective use or replacement of models with lower accuracy 'dummy' models based on application requirements, thereby reducing processing load and costs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a common simulation architecture integrates all simulation data models (vehicle simulator, sensor simulator, etc.) to ensure comprehensive testing capability, then testing reliability is improved, but device complexity and processing load increase

Engineering Contradiction:
Improvetesting reliabilityVSAvoidsimulation architecture complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The simulation architecture is segmented into independent, switchable model modules (vehicle simulator, sensor simulator, core simulator) that can be selectively activated. Each module operates independently and can be connected or disconnected based on testing requirements, reducing overall system complexity while maintaining reliability when needed.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The simulation architecture implements dynamic configurability where the connection status of each simulation model can be changed during operation. Switches allow the system to transition between different operational states (full integration, partial integration, or minimal integration) based on the specific testing needs of different applications.

Inventive Principle:
Principle #15Dynamics

2Measurement precision

If highly accurate sensor simulators are integrated into the simulation architecture to ensure data accuracy, then measurement precision is improved, but processing load and cost increase

Engineering Contradiction:
Improvesensor data accuracyVSAvoidprocessing load
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

Different levels of simulation accuracy are applied locally based on specific testing requirements. Instead of using high-accuracy models universally, the system allows selective activation of accurate sensor simulators only when and where precision is needed for particular applications, reducing overall processing load while maintaining necessary accuracy.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system enables dynamic changing of simulation parameters and model fidelity levels. Simulation models can be switched between different accuracy modes or completely disconnected based on the specific testing scenario, allowing optimization of processing load while maintaining measurement precision when required.

Inventive Principle:
Principle #35Parameter changes

3Adaptability or versatility

If all simulation data models are integrated into the core simulator to provide comprehensive testing capability, then adaptability is improved, but processing time increases

Engineering Contradiction:
Improvetesting capabilityVSAvoidprocessing time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system performs preliminary configuration by allowing users to select and pre-configure only the necessary simulation models before executing tests. By anticipating which models are needed for specific application types, the system avoids unnecessary processing of unrelated models, reducing processing time while maintaining adaptability.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The core simulator is designed with universal interfaces that can connect to different simulation models as needed. This multi-functional architecture allows the same core simulator to work with various combinations of models (vehicle only, sensor only, both, or neither) without requiring separate systems, thus maintaining adaptability while enabling selective activation to reduce processing time.

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

Data Source

PatentEP4242854A1System and method for providing autonomous driving simulation architecture with switchable models
Publication Date: 2023.09.13 WOVEN BY TOYOTA INC
  • EP4242854A1 patent drawingFigure 1
  • EP4242854A1 patent drawingFigure 2
  • EP4242854A1 patent drawingFigure 3

AI summary

A device and a method for providing an autonomous driving simulation architecture for testing an application. The method includes integrating a plurality of simulation data models with a first core simulator that runs an autonomous vehicle simulation for testing the application; selectively connecting the first core simulator to the application; and selectively disconnecting a first simulation data model, among the plurality of simulation data models, from the first core simulator.