Low-Impact Collision Scoring with High-Rate Vehicle Acceleration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for detecting vehicle collisions rely solely on accelerometer data, leading to unreliable results due to false positives and false negatives, as they cannot differentiate between collision-induced accelerations and normal vehicle movements or external factors like potholes, and lack contextual information for accurate collision characterization.
Innovation Solution
A system that captures high-rate acceleration data and combines it with GPS and speed data to compute an accident score, using feature values and scores to filter out noise and harsh braking events, and trigger appropriate actions upon determining a potential collision, thereby enhancing the reliability of collision detection and characterization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional methods use only accelerometer data for collision detection, then the system is simple to operate, but the detection reliability is low due to false positives and false negatives
Solution Approach 1:
The patent combines multiple data sources (accelerometer data, GPS data, speed data) into a unified collision detection system. The server integrates these diverse data types to compute an accident score, merging information from different sensors to improve detection reliability while distributing processing complexity between the vehicle device and server.
Solution Approach 2:
The system uses a multi-functional approach where the same data processing infrastructure handles multiple tasks: collision detection, accident score computation, and contextual analysis. The server performs multiple functions including receiving data from various sources, processing different data types, and generating comprehensive collision assessments.
2Measurement precision
If the system captures high-rate acceleration data and computes comprehensive accident scores, then the measurement precision is improved, but the loss of time for data processing increases
Solution Approach 1:
The system performs preliminary data collection and processing by capturing high-rate acceleration data continuously and pre-computing feature values from the raw data. The accident score computation uses pre-calculated features (area under curve, standard deviation, maximum acceleration) to reduce real-time processing requirements while maintaining high measurement precision.
Solution Approach 2:
The data processing is segmented into distinct stages: data capture, feature extraction, accident score computation, and collision determination. Each stage processes specific aspects of the data independently, allowing parallel processing and reducing overall processing time while maintaining comprehensive analysis precision.
3Reliability
If the system filters out noise events and harsh braking events, then the reliability is improved, but the device complexity increases
Solution Approach 1:
The system uses parameter-based filtering by computing specific features (area under curve, standard deviation, maximum acceleration) and comparing them against threshold values. Events are classified and filtered based on their computed parameters, allowing the system to distinguish between collision events and non-collision events like noise or harsh braking through quantitative parameter analysis.
Data Source
AI summary
Described herein are various techniques, including a method that uses high-rate acceleration data for computing an accident score indicative of a potential collision and triggering an action in response to determining that the accident score indicates a potential collision. The method includes filtering out undesired high-rate acceleration trigger events such as noise and harsh braking events prior to determining the accident score. The accident score is based on contexts or scores computed from high-rate acceleration data, speed, and GPS data captured by a telematics monitor deployed in a vehicle.