Vehicle Occupant Pose Control for Intuitive Actuator Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Passengers face difficulties in intuitively adjusting and controlling vehicle components using unnatural hand gestures, which are not directly related to the actions they trigger.
Innovation Solution
A method and device that capture images or videos of a person in the vehicle using an image capturing unit, process the data to generate a body data model, compare it with reference body poses stored in a database, and activate corresponding vehicle actuators based on matched poses, allowing intuitive control through natural body positions and orientations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If hand gestures are used to control vehicle components, then control capability is provided, but the gestures become unnatural and difficult to remember
Solution Approach 1:
The patent replaces the mechanical gesture recognition system with a vision-based system using cameras and image processing. Instead of requiring passengers to perform specific hand gestures, the system captures images of the passenger's body pose and automatically interprets the intended action based on the pose configuration, thereby substituting complex mechanical gesture control with optical detection and computational analysis
Solution Approach 2:
The system allows the passenger's natural body pose to automatically serve as the control input. The passenger simply positions their body naturally (e.g., leaning forward, turning head), and the system self-interprets this pose to trigger the corresponding vehicle component adjustment, eliminating the need for the passenger to learn or remember specific gestures
2Adaptability or versatility
If multiple hand gestures are defined for different actions, then control versatility is improved, but the learning burden increases
Solution Approach 1:
The system changes the control parameter from discrete hand gestures to continuous body pose parameters. By detecting key body points and calculating pose angles and positions, the system can distinguish between numerous different body configurations, each corresponding to different control actions. This parameter transformation enables versatile control without requiring memorization, as any natural body pose can potentially serve as a control input
3Ease of operation
If natural body poses are used for control, then ease of operation is improved, but image processing complexity increases
Solution Approach 1:
The patent segments the body into key detectable points (such as joints, head position, torso position) rather than attempting to analyze the entire body continuously. By identifying and tracking these discrete key points, the system simplifies the image processing task while still capturing sufficient information to determine body pose and intent, thereby reducing processing complexity while maintaining natural control
Data Source
AI summary
A method for adjusting or controlling at least one actuator in a vehicle (1), for example an electrical servo motor (2) for a seat (3), a tripping mechanism of an airbag (4), or a display device (5) in the vehicle (1), comprising the following steps: capturing an image or a video of a preferably sitting person in the vehicle (1) through an image capturing unit (6) in 2D or 3D arranged in the vehicle (1), extracting the body data of the person through a data processing unit (7) and generation of a body data model (8) from the body data through a data processing unit (7), comparing the body data model (8) with a plurality of reference body poses (11) obtained from a database (10) through a comparator unit (9) and selecting a corresponding reference body pose (11), activating an actuator assigned to the selected reference body pose (11) with a setting of the actuator assigned to the selected reference body pose (11) through an activation unit (12).