Adaptive EV Driving Sound Control Using Driver State Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electric vehicles lack personalized driving sounds that can satisfy drivers' desires for excitement and emotional connection, as they rely on pre-designed sounds that do not reflect the driver's physical condition.
Innovation Solution
A vehicle system equipped with a camera, sensor, and controller that determines the driver's physical condition through face images and skin information, generates a driving sound by combining sound sources and parameters based on the driver's state and vehicle information, using AI algorithms like SVM and Eulerian video magnification to estimate pulse rate and detect galvanic skin response.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If pre-designed driving sounds are used in electric vehicles, then the vehicle can provide basic driving feedback, but the sounds cannot satisfy drivers' personalized demands for excitement and emotional connection
Solution Approach 1:
The system dynamically adjusts driving sound parameters in real-time based on detected driver physical conditions. The controller continuously monitors physiological data and modifies sound characteristics accordingly, transforming the static pre-designed sound system into a dynamic, adaptive audio experience that responds to driver state changes.
Solution Approach 2:
The system changes multiple sound parameters including frequency, amplitude, waveform type, and timbre based on driver physical conditions. By adjusting these acoustic parameters in response to detected physiological states, the system provides personalized driving sounds without requiring completely different sound generation hardware.
2Adaptability or versatility
If AI algorithms and multiple sensors are added to determine driver physical condition, then personalized driving sounds can be generated, but the device complexity and processing requirements increase
Solution Approach 1:
The camera system performs multiple functions: capturing driver facial images for emotional state analysis, detecting blood flow changes for physiological monitoring, and providing basic driver presence detection. This multi-functional use of existing hardware reduces the need for additional specialized sensors while achieving comprehensive driver state recognition.
Solution Approach 2:
The system uses facial blood flow patterns as an intermediary indicator to infer deeper physiological states. Instead of directly measuring complex physiological parameters, the system detects changes in facial blood flow through camera-based photoplethysmography, which serves as a non-invasive mediator to estimate driver arousal, stress, and emotional states.
3Measurement precision
If real-time processing of face images and skin information is performed, then accurate driver physical condition determination is achieved, but the processing time and computational load increase
Solution Approach 1:
The system pre-processes and stores baseline driver facial characteristics and emotional state patterns during calibration phases. By having reference data ready in advance, the real-time detection process can quickly compare current readings against pre-established patterns, reducing computational complexity and processing time during actual driving operations.
Solution Approach 2:
The processing pipeline is divided into separate stages: initial facial feature extraction, blood flow pattern detection, physiological state estimation, and sound parameter selection. Each segment processes specific aspects of driver state independently, allowing parallel processing and reducing overall computation time compared to holistic analysis approaches.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Provides a personalized driving experience by generating sounds that match the driver's emotional and physical state, enhancing driving pleasure and marketability, especially in shared mobility scenarios where various sounds can be tailored to different drivers.
Implementation Method 1
The controller may be configured to apply an Eulerian video magnification framework to the face image, extract a blood flow image from the face image
Implementation Method 2
The skin information may include a galvanic skin response (GSR)
Data Source
AI summary
A vehicle is provided and includes a camera that photographs a face of a driver, a sensor that obtains skin information of the driver, a speaker, and a controller. The controller determines a physical condition of the driver based on at least one of a face image of the driver and the skin information of the driver, determines a sound source and a parameter based on the physical condition of the driver, generates a driving sound by reflecting driving information of the vehicle in the sound source and the parameter, and operates the speaker to output the driving sound.


