Vehicle Learning Control Stop Logic for Outdated Parts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle control systems using machine learning can perform unnecessary control operations if a vehicle is equipped with parts scheduled for repair, replacement, or outdated parts, leading to wasteful use of processing resources and power.
Innovation Solution
A server communicates with vehicles to identify if they are equipped with parts that need repair, replacement, or are outdated, and sends instructions to stop machine learning-related control operations in such cases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If learning related control is performed in vehicles with parts scheduled for repair, replacement, or outdated parts, then machine learning training can be conducted, but the control becomes wasteful and consumes unnecessary processing resources and power
Solution Approach 1:
The server performs preliminary checking of part status information before allowing learning related control to proceed. By advance identifying vehicles with repair-scheduled parts, replacement-scheduled parts, or outdated parts, the system prevents wasteful machine learning training from occurring in the first place, thereby conserving processing resources and power while maintaining training productivity in eligible vehicles
Solution Approach 2:
The system establishes a feedback mechanism where the server continuously monitors part status information from vehicles and provides control instructions back to vehicles. Based on the feedback regarding part conditions (repair scheduled, replacement scheduled, or outdated), the server dynamically adjusts whether learning related control should be executed, ensuring that processing resources are only consumed when training can be effectively performed
2Loss of information
If learning related control is performed in vehicles with outdated parts, then training data can be collected, but the training results may be inaccurate or obsolete
Solution Approach 1:
The server performs preliminary verification of part status before allowing learning control execution. By checking whether parts are outdated in advance, the system prevents collection of low-quality training data from vehicles with obsolete components, thereby maintaining training data quality without unnecessarily reducing learning control productivity in vehicles with current, reliable parts
Data Source
AI summary
A model learning system comprises a server and a plurality of vehicles having control devices configured to be able to communicate with the server and configured to perform learning related control relating to machine learning. The server is configured to judge if among the plurality of vehicles, there is a specific vehicle mounting a repair part scheduled for repair, a replacement part scheduled for replacement, or an outdated part and, when there is a specific vehicle, to transmit an instruction to the specific vehicle to stop learning related control. The control device is configured so that, when receiving a stop instruction, it makes the learning related control stop.