On-Board Diagnostics Scheduling Using Learned Road Segment Windows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle diagnostic systems face challenges in maintaining high in-use monitor performance (IUMP) rates due to unfavorable conditions such as engine operation transitions and noisy environments, leading to incomplete diagnostic routines and decreased compliance with regulatory mandates.
Innovation Solution
A method that learns the characteristics of a road segment during travel, allowing for the scheduling of diagnostic routines based on statistical parameters, including boosted and naturally aspirated engine operations, and noise factors, to predict favorable windows for routine completion and adjust the Z-score for improved IUMP rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If diagnostic routines are executed during boosted engine operations, then diagnostic coverage is improved, but transition to naturally aspirated operation during routine execution causes inaccuracy
Solution Approach 1:
The system performs preliminary characterization of road segments during previous travels, learning statistical parameters including boosted and naturally aspirated engine operation events. This advance knowledge allows the diagnostic routine to be scheduled during favorable windows where the engine will remain in boosted operation throughout the entire routine execution, preventing transitions that would cause inaccuracy.
Solution Approach 2:
The system dynamically adjusts the diagnostic scheduling strategy based on learned road segment characteristics. By using statistical parameters and Z-scores to predict engine operation states, the system adapts the timing and placement of diagnostic routines to match actual engine behavior patterns, ensuring routines execute during stable boosted conditions rather than during transitions.
2Productivity
If diagnostic routines are scheduled based on entry conditions, then routine execution is simplified, but unfavorable conditions such as noisy environments cause incomplete routines and decreased IUMP rates
Solution Approach 1:
The system performs preliminary learning of road segment characteristics including noise factors such as fuel sloshing and bumpy road conditions during previous travels. This advance characterization allows the system to identify and avoid unfavorable windows where noise would interfere with diagnostic routine completion, scheduling routines only during quiet, stable operating conditions.
Solution Approach 2:
The system uses feedback from completed diagnostic routines to adjust the Z-score and refine future scheduling decisions. By monitoring whether routines complete successfully or are interrupted by noise factors, the system learns from past performance and improves its prediction of favorable execution windows, progressively increasing IUMP rates.
3Reliability
If diagnostic routines are carried out during noisy environments, then diagnostic coverage is maintained, but routine accuracy deteriorates and IUMP rates decrease
Solution Approach 1:
The system performs preliminary characterization of road segments, learning statistical parameters including noise factors such as fuel sloshing and bumpy road conditions during previous travels. This advance knowledge allows the diagnostic routine to be scheduled during favorable windows where noise factors are minimized, ensuring accurate and complete routine execution.
Solution Approach 2:
The system changes the scheduling parameters based on the learned statistical characteristics of each road segment. By using mean, variance, and Z-scores derived from historical data, the system identifies optimal time windows where engine operating conditions and noise levels are most favorable for diagnostic routine execution, thereby maintaining high completion accuracy.
4Measurement precision
If statistical parameters are learned during previous travels, then diagnostic scheduling accuracy is improved, but system complexity increases
Solution Approach 1:
The system performs self-learning by automatically characterizing road segments during normal vehicle operation. No external calibration or manual input is required - the system collects data during previous travels, computes statistical parameters, and uses this knowledge to improve future diagnostic scheduling autonomously, reducing the need for complex external configuration systems.
Solution Approach 2:
The system focuses on learning and storing only the essential statistical parameters (mean, variance, and event counts) needed for predictive scheduling, rather than attempting to model all possible road segment characteristics. This selective parameter approach achieves sufficient measurement precision for diagnostic scheduling while keeping the data structure and processing relatively simple.
Data Source
AI summary
Methods and systems are provided for learning characteristics of a road segment and using the learning during a subsequent travel on the road segment to preview the road segment and schedule diagnostic routines to increase in-use monitor performance (IUMP) rates for vehicle system diagnostics. In one example, a method may include scheduling a diagnostic routine for an engine sub-system based on the learned characteristics of the road segment, and a variable adjustable based on IUMP rate.