Mobile Vehicle Accident Detection Through Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting accidents require human intervention, which can lead to inaccurate information, prolonged response times, and inefficient resource management, especially when the individual is unconscious.
Innovation Solution
A method using a mobile device with sensors and AI algorithms to autonomously detect accidents by analyzing geolocation, vehicle dynamics, sound, and vital signs, and sending alerts to rescue services.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If passive methods relying on individual witnessing are used to detect accidents, then the system complexity is low, but the detection reliability and response time are poor
Solution Approach 1:
The mobile device automatically detects accidents using its own sensors (accelerometer, GPS, microphone) without requiring external witnesses or manual intervention. The device self-monitors for abnormal patterns indicating accidents and autonomously generates alerts, eliminating dependency on human witnesses while maintaining system simplicity.
Solution Approach 2:
The mobile device performs multiple functions: it serves as both the detection device (using built-in sensors) and the communication device (sending alerts). This multi-functionality eliminates the need for separate specialized detection equipment, improving reliability while avoiding increased system complexity.
2Speed
If manual alerting by witnesses is used, then the device complexity is low, but the response time is prolonged
Solution Approach 1:
The system continuously monitors accident parameters in real-time before an accident occurs. When abnormal patterns are detected, the alert is immediately generated and sent automatically, eliminating the time delay associated with manual witness detection and reporting. The system is already in a state of readiness to detect and report.
Solution Approach 2:
The manual mechanical process of witness observation and phone calling is replaced by automated electronic sensor detection and electronic alert transmission. The accelerometer, GPS, and microphone sensors automatically detect physical anomalies, and the communication module automatically sends alerts, dramatically reducing response time.
3Measurement precision
If information is provided by involved individuals, then no additional detection devices are needed, but the information accuracy and detail are insufficient
Solution Approach 1:
The mobile device automatically collects and processes accident data using its built-in sensors (accelerometer for impact detection, GPS for location, microphone for sound analysis). This self-service capability provides accurate, detailed objective information about the accident without relying on human memory or reporting ability.
Solution Approach 2:
The sensors act as intermediaries between the physical accident event and the digital information system. The accelerometer translates physical impact forces into measurable data, the GPS translates location into coordinates, and the microphone translates sounds into audio signals, providing precise objective information that bridges the physical and digital domains.
4Loss of information
If unconscious individuals are relied upon for information, then the system remains simple, but no information can be provided at all
Solution Approach 1:
The mobile device independently performs all detection and reporting functions without requiring the user to be conscious or able to interact. The sensors continuously monitor for accident conditions and automatically generate and send alerts, ensuring information is provided even when the individual is unconscious.
Solution Approach 2:
The system is pre-configured to automatically detect and report accidents without requiring user activation or consciousness. The detection algorithms are already running in the background, monitoring sensor data for accident patterns, so when an accident occurs, the system immediately acts without needing the user's input.
Data Source
Figure 1
Figure 2
AI summary
Method for detecting an accident (1) of a vehicle (100) carried out by a mobile device (10) associated with a user in the vehicle, comprising the steps of: a) capturing a first dataset (Set1) by first geolocalization sensors (S1) embedded in the mobile device (10) in sampling time intervals; b) calculating a vehicle cruising speed (Vc) by the first dataset (Set1) in each sampling time interval; c) identifying an abrupt braking (Fb) in case a comparison between cruising speeds (Vc) in temporally successive sampling intervals exceeds first predefined threshold values (Vs1); d) if an abrupt braking (Fb) is identified, capturing a second dataset (Set2) by second sensors (S2) embedded in the device (10), and identifying an accident condition (Ci) by comparing values associated with the second dataset (Set2) and second predefined threshold values (Vs2), said second dataset (Set2) being representative of the conditions inside the vehicle; e) if an accident condition (Ci) is identified, generating a first accident signal (Si1); f) sending the first accident signal (Si1) from the mobile device (10) associated with the user to one or more external devices (20) placed in signal communication with each other, at least one external device (20) being associated with the rescue services.