Vehicle Image Analysis Focus-of-Expansion Learning Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Error learning of a focus-of-expansion position occurs during simulated vehicle runs, where features like stains on walls or shadows are mistakenly identified as road division lines, leading to incorrect estimation.
Innovation Solution
An image analysis apparatus with a camera, learning means, and controlling means that only starts learning the focus-of-expansion position if the vehicle's speed exceeds a reference speed for a predetermined duration, preventing learning during simulated runs on a chassis dynamometer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If focus-of-expansion learning is performed continuously, then learning accuracy improves, but error learning occurs during simulated runs
Solution Approach 1:
The system changes the operational parameter of the learning function by enabling or disabling it based on vehicle speed conditions. When vehicle speed exceeds a predetermined threshold, learning is enabled; when speed is below the threshold, learning is disabled. This parameter change prevents error learning during simulated runs while maintaining learning accuracy during normal driving conditions.
Solution Approach 2:
The vehicle speed detection function serves as an intermediary that mediates between the learning function and the actual driving conditions. By using speed as an intermediate indicator, the system can infer whether the vehicle is in a simulated run state and accordingly control the learning process, preventing error learning without directly detecting simulated run conditions.
2Quantity of substance
If focus-of-expansion learning is performed during all vehicle operations, then more data is collected for learning, but incorrect features are learned from simulated run environments
Solution Approach 1:
The system dynamically changes the learning function's operational state based on vehicle speed parameters. By switching the learning function on or off according to speed thresholds, the system ensures that only data from actual road conditions (where speed exceeds the threshold) is used for learning, thereby maintaining both data quantity and learning precision.
Solution Approach 2:
The system applies different learning qualities to different operating conditions. During simulated runs (low speed), learning is disabled with zero quality weight, while during normal driving (high speed), learning is enabled with full quality weight. This local differentiation ensures that only high-quality data from appropriate conditions contributes to the learned model.
Data Source
AI summary
An image analysis apparatus picks up an image of a region ahead of a vehicle using a camera, and allows a control unit to analyze picked-up image data generated by the camera to learn a focus-of-expansion position. The control unit controls the learning performance as follows. Specifically, the control unit does not start the learning performance for the focus-of-expansion position until a state where a detection value of a vehicle speed exceeds a reference speed exceeds a specified duration of time. When the state where a detection value of a vehicle speed exceeds a reference speed exceeds the specified duration of time, the learning performance for the focus-of-expansion position is started from this time point. The specified duration of time may be determined on the basis of statistics on the durations of simulated runs of the vehicle performed on a chassis dynamometer in a vehicle inspection.