Vehicle Control Notification for Machine Learning Transparency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Operators of vehicles may erroneously recognize whether control functions using machine learning models are employed, leading to potential unexpected situations due to varying usage situations of machine learning models across different control functions.
Innovation Solution
A control system in vehicles that determines whether a machine learning model is used in executing control functions and notifies the operator when such a model is employed, utilizing processors to execute processes and provide notifications through displays, sounds, or actuator feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If machine learning models are used in control functions, then the control capability and adaptability are improved, but the operator's understanding and control over the system deteriorate due to erroneous recognition
Solution Approach 1:
The system implements feedback by notifying the operator when a machine learning model is being used for control. The notification unit provides real-time information about AI model usage, allowing the operator to understand the current control state and adjust accordingly, thus preventing erroneous recognition while maintaining ML model benefits
Solution Approach 2:
The notification unit acts as an intermediary between the machine learning model and the operator. It translates the internal AI model usage state into visible information for the operator, bridging the knowledge gap without requiring the operator to understand complex ML operations
2Extent of automation
If machine learning models are used in control functions, then the automation level is improved, but the transparency and traceability of control decisions deteriorate
Solution Approach 1:
The notification unit provides feedback to the operator about when AI models are making control decisions. This maintains automation while improving transparency by making the invisible AI decision-making process visible and understandable to the human operator
Data Source
AI summary
A control system installed in a vehicle comprises one or more processors. The one or more processors are configured to execute one or more control functions in response to a driving environment of the vehicle or a request. The one or more processors determines whether or not a machine learning model is used in a control function being executed. And when it is determined that the machine learning model is used, the one or more processors notify an operator of the vehicle that the control function using the machine learning model is being executed.