Multi-Layer Grid Proximity Detection for Vehicle Safety
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems lack effective automated methods to detect and analyze near-incident situations involving multiple vehicles, leading to undetected hazardous driving behaviors and increased road safety risks due to the absence of autonomous systems that can identify suspicious dangerous situations.
Innovation Solution
A method and apparatus that utilize motion sensors from multiple vehicles to collect and analyze data, storing it in a multi-layer grid database to determine vehicle proximity and identify potentially risky maneuvers, enabling the detection of dangerous scenarios through 'what-if' simulations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual analysis of accident scenarios is used to identify driver behavior, then detailed understanding of specific incidents can be achieved, but the process requires tedious reconstruction of the scene and cannot detect the majority of near-incident events
Solution Approach 1:
The patent replaces manual mechanical analysis with automated sensor-based detection systems. Motion sensors (accelerometers, gyroscopes) automatically capture vehicle movement data, and algorithms process this data to identify dangerous maneuvers without human intervention, thereby detecting the majority of near-incident events that manual analysis would miss
Solution Approach 2:
The patent introduces an intermediary automated analysis system that acts as a mediator between raw sensor data and safety conclusions. This system processes motion sensor data, identifies hazardous patterns, and generates safety assessments, enabling both precise detection and broad coverage of near-incident events
2Extent of automation
If motion sensors are integrated into vehicles to collect data, then automated detection of dangerous maneuvers becomes possible, but the data must be interpreted relative to the sensor's physical orientation which complicates analysis
Solution Approach 1:
The patent transforms the coordinate reference frame from sensor-relative to vehicle-relative by applying rotation matrices that convert accelerometer and gyroscope data into the vehicle's coordinate system. This parameter transformation simplifies the interpretation of motion data by aligning it with the vehicle's physical orientation rather than the sensor's arbitrary orientation
Solution Approach 2:
The patent introduces coordinate transformation algorithms as an intermediary layer between raw sensor data and meaningful motion interpretation. These algorithms act as a mediator that converts sensor-relative measurements into vehicle-relative measurements, reducing the complexity of subsequent analysis
Data Source
Figure 1a~2b
Figure 3
Figure 4
AI summary
There are disclosed various methods and apparatuses for multi-vehicle manoeuvre and impact analyses. According to an embodiment, the apparatus comprises means for: receiving sensor data from at least one motion sensor associated with a selected vehicle from a plurality of vehicles, said sensor data indicating movements of the vehicle in a geographical area; obtaining from the sensor data location of the vehicle in the geographical area; maintaining a vehicle location and motion database having at least three grid layers of grid cells representing at least a part of the geographical area, said grid cells of the at least three grid layers being partly overlapping so that borders of the grid cells of different grid layers have an offset between each other; mapping the location of the vehicle to a grid cell of each grid layer into which the location of the vehicle belongs; inserting or maintaining an identifier of the vehicle in a list of vehicle identifiers for the mapped grid cell of each grid layer; and determining on the basis of the identifiers of the vehicles in the lists which other vehicles are in a proximity of said selected vehicle.