Mobile Device Accident Detection Using Multi-Level Sensor Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle accident detection methods are unreliable due to the reliance on short-range connections between mobile phones and vehicles, often resulting in incorrect accident reports and warnings, which can lead to an overload of control centers.
Innovation Solution
A method utilizing a mobile device equipped with multiple sensors to continuously record and store sensor data, evaluate significance and credibility levels, and autonomously decide on sending accident reports, ensuring reliable detection and evaluation of accidents without external device dependencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a vehicle emergency sensor sends a trigger signal to an emergency control unit and a mobile phone receives the emergency signal via short-range communication, then an emergency call can be transmitted to an external control center, but the system reliability deteriorates because establishing the short-range connection is often forgotten if the mobile phone is not already configured for short-range connections
Solution Approach 1:
The mobile device autonomously performs accident detection and report generation without requiring external vehicle system configuration or user intervention. The device uses its own sensors (accelerometer, gyroscope, microphone) to detect accidents and automatically creates and sends accident reports, making the system self-sufficient and eliminating dependency on vehicle-integrated emergency systems
Solution Approach 2:
The accident detection and reporting functionality is extracted from the vehicle's emergency control system and implemented independently on a portable mobile device. This separation allows the mobile device to operate autonomously without requiring short-range communication setup with the vehicle, while still achieving the same emergency notification goal
2Reliability
If sensor data from multiple sensors are continuously recorded and stored in memory, then accident detection reliability is improved through multi-level significance and confidence level evaluation, but the device complexity increases due to the accident verification routine and multiple sensor integration
Solution Approach 1:
The accident verification process is segmented into distinct evaluation stages: significance level determination (evaluating individual sensor readings) and confidence level determination (evaluating the combination of multiple sensor readings). This segmentation allows the complex verification routine to be broken down into manageable, modular components that can be independently optimized and maintained
Solution Approach 2:
The system uses multiple sensors measuring different physical parameters (acceleration, angular velocity, sound pressure) to detect accidents. By changing the parameters being measured and combining them through multi-level evaluation, the system achieves high reliability without requiring a single overly complex sensor or algorithm
3Reliability
If the accident verification routine repeatedly determines significance level and confidence level within a specified time interval, then false accident reports are reduced through differentiated verification, but the processing time increases due to repeated evaluations and decisions
Solution Approach 1:
The accident verification routine performs repeated significance and confidence level determinations at periodic intervals within a specified time window after an accident is detected. This periodic verification allows the system to monitor the situation over time and confirm whether the accident condition persists or was a false alarm, balancing accuracy with reasonable processing time
Solution Approach 2:
The system uses feedback from repeated significance and confidence level evaluations to dynamically adjust its decision-making. Each evaluation provides feedback about the current state, and the accumulation of feedback over multiple evaluations enables the system to make more reliable decisions while avoiding premature or false accident reports
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
This approach significantly reduces incorrect accident reports and warnings, enhancing the reliability of accident detection and preventing system overloads by using multi-level significance and credibility evaluations on a single mobile device.
Implementation Method 1
a first sensor, in particular an accelerometer
Implementation Method 2
a second sensor, in particular a microphone
Implementation Method 3
a third sensor, in particular a speed sensor
Data Source
AI summary
The invention relates to a method for detecting and evaluating a vehicle accident, wherein the method steps are carried out on a mobile device equipped with at least two sensors, which is carried with the vehicle.In this process, an accident review routine is executed in such a way that the stored sensor data is evaluated to determine a multi-level significance level (severity level), and that the sensor data recorded from the timestamp onwards within a specified time interval are also evaluated to repeatedly determine a multi-level confidence level, and that within this time interval, based on the significance level (severity level) and/or the currently determined confidence level, a decision is repeatedly made as to either submit an accident report, or not submit an accident report and the accident review routine is executed again, or the accident review routine is aborted.