Autonomous Driving HMI Feedback for Sensor Degradation Risk
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Machine learning models in autonomous driving systems may not adequately handle all driving scenarios due to varying traffic environments, necessitating effective monitoring of vehicle control.
Innovation Solution
An autonomous driving system that utilizes sensors to detect vehicle and surrounding conditions, implements vehicle control via a machine learning model, and presents detection results through a Human Machine Interface (HMI) when sensor performance is below a predetermined level or a risky situation occurs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If machine learning models are used for vehicle control, then automation capability is improved, but reliability deteriorates due to inability to handle all driving scenarios
Solution Approach 1:
The system continuously monitors sensor detection performance during autonomous driving and provides feedback when performance degrades below predetermined levels or when risky situations occur. This feedback mechanism enables real-time assessment of ML model performance and triggers appropriate responses such as presenting detection results to the driver or switching to manual control, thereby maintaining reliability while preserving automation benefits
Solution Approach 2:
The system performs preliminary monitoring of sensor detection performance before critical failures occur. By continuously evaluating detection performance metrics and comparing them against predetermined thresholds, the system proactively identifies degradation trends and takes preventive actions (such as alerting the driver or switching control modes) before the ML model's limitations compromise safety
2Reliability
If sensor detection performance monitoring is implemented, then reliability is improved, but device complexity increases
Solution Approach 1:
The system monitors its own sensor detection performance autonomously without requiring external monitoring infrastructure. The autonomous driving ECU evaluates detection performance metrics from sensor data itself, determining when performance degradation occurs and when risky situations arise, thereby improving reliability through self-monitoring while avoiding the complexity of additional external monitoring systems
Data Source
AI summary
A autonomous driving system detects information related to at least one of driving conditions and surrounding conditions of a vehicle using one or more sensors, implements vehicle control using a machine learning model based on the information, presents the detection results of the sensor when a detection performance of the sensor is lower than a predetermined level, and presents the detection results of the sensor corresponding to a situation when a situation with a risk higher than a specified value occurs in the vehicle.