Pneumatic Brake Self-Diagnosis Using Learned Response Baselines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronically controlled pneumatic braking systems in autonomous vehicles lack effective methods for detecting faults, particularly those that occur infrequently and cannot be sensed by direct sensors, and do not account for manufacturing or age-related variations.
Innovation Solution
A diagnostic method that involves putting the braking system into a learning mode, performing predefined activities, and capturing system responses using existing sensors to establish a reference, which is then compared against stored target responses to detect deviations and output fault signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a driver-dependent diagnostic approach is used, then the system structure remains simple, but the system cannot reliably detect faults in autonomous vehicles without a driver
Solution Approach 1:
The braking system performs self-diagnosis by automatically comparing its own actual responses against stored target responses. The control unit executes predefined activities, captures sensor data, and autonomously determines deviations without driver intervention, enabling the system to monitor itself reliably in autonomous operation modes.
Solution Approach 2:
Target system responses are pre-stored in the control unit for various operating conditions. These reference values are established beforehand and used during operation to quickly compare against actual system responses, enabling immediate fault detection without requiring complex real-time analysis models.
2Measurement precision
If fixed threshold values are used for fault detection, then the diagnostic procedure is simple, but manufacturing and age-related variations cause false positives
Solution Approach 1:
Instead of fixed threshold values, the system uses dynamic comparison against stored target responses that represent actual system behavior under specific conditions. The diagnostic approach adapts to manufacturing variations and aging by comparing against empirically determined reference values rather than rigid thresholds, improving detection accuracy.
Solution Approach 2:
The system continuously monitors actual system responses and compares them against target responses, providing feedback on deviations. This feedback mechanism allows the system to identify faults based on behavioral changes rather than fixed limits, accommodating natural variations in system performance over time and across manufacturing batches.
3Reliability
If the vehicle is driven under specific conditions for diagnostics, then diagnostic data can be collected, but the driver's operation is interrupted and productivity decreases
Solution Approach 1:
The diagnostic procedure operates continuously during normal vehicle operation without requiring separate diagnostic modes or interrupting the driver. The control unit performs comparisons between actual and target responses in real-time, allowing diagnostics to proceed continuously alongside normal driving tasks, maintaining both diagnostic reliability and operational productivity.
Solution Approach 2:
The braking system serves dual functions: normal braking operation and self-diagnosis. The same control unit and sensor arrangements used for operational control are also utilized for diagnostic purposes, eliminating the need for separate diagnostic hardware or dedicated diagnostic driving conditions, thereby maintaining productivity while ensuring diagnostic quality.
4Reliability
If additional sensors are added to detect all faults, then fault detection coverage improves, but system complexity and cost increase
Solution Approach 1:
The system uses existing sensor arrangements to monitor its own performance by capturing actual system responses during operation. Rather than adding dedicated fault detection sensors, the system repurposes existing sensors to provide diagnostic data, comparing actual responses against target responses to identify faults, thereby improving detection coverage without increasing hardware complexity.
Solution Approach 2:
Existing sensors serve dual purposes: monitoring system parameters for normal operation and providing diagnostic data for fault detection. The sensor arrangements are utilized for both operational control and self-diagnosis, eliminating the need for additional dedicated diagnostic sensors and maintaining system simplicity while improving fault detection coverage.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a diagnostic method (2) for performing a self-diagnosis of an electronically controllable pneumatic braking system (1) for a utility vehicle (4), said method comprising the steps of: - receiving a learning signal (SL) at the braking system (1); - in response to said signal being received, placing the braking system in a learning mode (102) and performing the steps: - executing a predefined first activity (104) of the braking system (1) while the utility vehicle (4) is stationary or moving; - using a sensor assembly (108) to detect a first learning system response (106) of the braking system (1) in response to the execution of the first activity (104); and - storing (114) the detected first learning system response (106) as a first target system response (107, 210) in a storage unit (82). The invention also relates to a braking system (1) and to a computer programme.