Steering Wheel Hand Detection for Autonomous Activation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semi-autonomous and autonomous driving systems require manual activation and button pressing, which can be inconvenient and may not accurately detect when a driver has removed their hands from the steering wheel, potentially leading to unsafe transitions between driving modes.
Innovation Solution
An activation system using an imaging means, such as a camera, to monitor the driver's hands on the steering wheel, processing image data to determine if both hands are off the wheel, and automatically triggering semi-autonomous or autonomous control systems without further driver input, by comparing raw images with background data sets and adjusting brightness thresholds to identify hand presence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual activation with button pressing is used, then the system requires simple hardware components, but the activation process becomes inconvenient and may not accurately detect driver intent
Solution Approach 1:
The patent replaces manual button pressing (mechanical input) with an optical detection system using imaging means to detect driver hand position. This substitution enables automatic activation based on visual recognition of hand-off-steering-wheel gesture, improving ease of operation while the control module processes image data to trigger the system
Solution Approach 2:
The system automatically detects driver intent through imaging means and triggers activation without requiring manual button pressing. The driver simply needs to perform the hand-off gesture, and the system self-activates based on image processing and pattern recognition algorithms
2Measurement precision
If imaging means is used to detect hand position, then activation accuracy improves, but the device complexity increases
Solution Approach 1:
The patent introduces an image processing pipeline with multiple intermediate processing steps including background subtraction, thresholding, and connected component analysis. These intermediary processing stages transform raw image data into reliable hand position detection, achieving high measurement precision through systematic data transformation
Solution Approach 2:
The detection process is segmented into distinct processing stages: background image creation, real-time image comparison, threshold application, and hand region identification. This segmentation of the image processing task enables complex detection functionality to be implemented through modular, manageable processing steps
3Ease of operation
If automatic triggering is implemented, then driver convenience is improved, but false activation risk increases
Solution Approach 1:
The system performs preliminary actions by continuously capturing and processing images in real-time, maintaining a background image for comparison, and pre-processing image data through thresholding and region analysis. This continuous preliminary processing ensures that when the hand-off gesture occurs, the system is already prepared to accurately detect and respond to the gesture without false triggers
Solution Approach 2:
The system uses feedback from continuous image analysis to verify hand position and determine whether activation conditions are met. By comparing real-time images against background images and analyzing hand region characteristics, the system provides feedback validation to confirm genuine hand-off gestures before triggering activation, reducing false positives
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to an activation system for a vehicle comprising a steering wheel (20) and a semi-autonomous or an autonomous control system. The activation system comprises at least one imaging means (12), such as a camera, having a field of view encompassing at least a substantial portion of the steering wheel arranged to collect, in use, a first image data set representing a raw image of a subject within said field of view. A control module is also provided for processing said first image data and for making a determination as to whether a driver currently has both of their hands on the steering wheel or not. In response to making said determination, the activation system is configured and/or arranged to trigger a semi-autonomous or an autonomous control system.