Automated Driving Control Logic Verification Using Behavior Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automated driving technologies face challenges in verifying control logic due to data conversion into discrete formats, leading to insufficient verification and potential loss of information, which affects the accuracy and reliability of automated driving systems.
Innovation Solution
A vehicle control device and system that includes detectors for vehicle behavior and peripheral situations, a storage for control logic, and a determinator to compare actual vehicle behavior with derived behavior based on detected situations, ensuring accurate operation of control logic and enabling detailed verification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If driving data is converted into discrete data to reduce communication data amount, then communication efficiency is improved, but information completeness deteriorates leading to insufficient control logic verification
Solution Approach 1:
The patent extracts only the essential verification results (match/mismatch determination) and key behavioral parameters from the continuous driving data, converting them into discrete verification data. This extraction process removes unnecessary detailed information while retaining the critical elements needed for control logic verification, thereby reducing communication data amount without significantly compromising verification effectiveness.
Solution Approach 2:
The patent changes the parameter representation from continuous driving data parameters to discrete verification result parameters. By transforming the data format and parameter types during the verification process, the system achieves efficient discrete data communication while maintaining the ability to accurately verify control logic through the determinator's comparison functions.
2Measurement precision
If control logic verification is performed in detail using continuous driving data, then verification accuracy is improved, but communication data amount increases
Solution Approach 1:
The patent extracts only the essential verification results (match/mismatch determination) and key behavioral parameters from the continuous driving data, converting them into discrete verification data. This extraction process removes unnecessary detailed information while retaining the critical elements needed for control logic verification, thereby reducing communication data amount without significantly compromising verification effectiveness.
Solution Approach 2:
The patent performs partial verification by focusing on key behavioral parameters and critical comparison points rather than processing all continuous driving data in detail. The determinator selectively verifies essential control logic functions through targeted comparison of derived behaviors against actual vehicle behaviors, achieving sufficient verification accuracy with reduced data processing and communication requirements.
3Reliability
If multiple control logics are verified simultaneously to improve system reliability, then verification comprehensiveness is improved, but computational complexity increases
Solution Approach 1:
The patent segments the verification process into distinct functional modules: the deriver executes individual control logics separately to derive expected behaviors, and the determinator compares each derived behavior against corresponding actual vehicle behaviors. This segmentation allows multiple control logics to be verified systematically through modular processing, improving comprehensiveness while managing complexity through structured division of verification tasks.
Solution Approach 2:
The patent creates virtual copies of control logic execution environments where each control logic is tested independently through the deriver. By simulating and comparing virtual derived behaviors against real vehicle behaviors without requiring simultaneous complex interactions, the system verifies multiple control logics thoroughly while keeping the verification process manageable through isolated testing copies.
Data Source
AI summary
A vehicle control device includes a first detector configured to detect a behavior of a vehicle, a second detector configured to detect a peripheral situation of the vehicle, a storage configured to store a control logic that outputs information indicating a behavior of the vehicle when at least information indicating a peripheral situation of the vehicle is input to the control logic, a deriver configured to execute the control logic to derive a behavior of the vehicle based on at least the peripheral situation of the vehicle detected by the second detector, and a determinator configured to compare a first behavior of the vehicle detected by the first detector at a certain reference time with a second behavior of the vehicle derived by the deriver based on the peripheral situation detected by the second detector at the reference time to determine whether or not the control logic is accurately operated.


