Vehicle Intersection Collision Avoidance via Predicted Path Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional collision alarm devices for vehicles at intersections often provide unreliable alerts due to incorrect determination of collision possibilities, either falsely alarming drivers about low-probability collisions or failing to alert them about high-probability collisions at a distance.
Innovation Solution
A vehicle control device that predicts the progress path of objects using pre-stored scenarios and determines collision possibilities based on the predicted paths, incorporating sensors like radar and cameras to set a variable region of interest and calculate Time-To-Collision (TTC) for targeted vehicles, enabling appropriate collision avoidance control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a conventional collision alarm device provides alarm based on proximity between sensed object and subject vehicle, then the alarm is provided for close objects, but the alarm reliability deteriorates due to false alarms for low collision probability scenarios and missed alarms for high collision probability scenarios at long distance
Solution Approach 1:
The patent changes the determination parameters from simple proximity-based alarm to a comprehensive collision possibility determination that includes progress path prediction, scenario matching, and TTC calculation. This transforms the alarm system from reacting to current distance to predicting future collision risk based on multiple parameters including object trajectories, intersection geometry, and vehicle dynamics
Solution Approach 2:
The system performs preliminary progress path prediction and scenario matching before determining collision possibility. By predicting where objects will be in the future and matching against pre-stored intersection scenarios, the system identifies potential collision risks in advance, enabling alarms for long-distance high-risk situations while filtering out short-distance low-risk situations
2Measurement precision
If the vehicle control device predicts progress path using multiple scenarios and determines collision possibility, then the collision determination accuracy is improved, but the device complexity increases
Solution Approach 1:
The patent segments the collision determination process into distinct functional modules: progress path prediction unit, scenario matching unit, TTC calculation unit, and collision possibility determination unit. Each module handles a specific aspect of the analysis, making the complex overall system manageable through functional decomposition and parallel processing
Solution Approach 2:
The system dynamically adjusts the region of interest and sensing parameters based on predicted progress paths and identified scenarios. Rather than uniformly monitoring all areas, the system focuses computational resources on dynamically identified high-risk zones and objects, optimizing the balance between comprehensive monitoring and computational efficiency
3Area of stationary object
If the vehicle sensor monitors a wide area to detect all objects, then the detection coverage is improved, but the sensing performance for target vehicles deteriorates due to distributed attention across multiple objects
Solution Approach 1:
The patent applies different sensing strategies to different spatial regions. The system identifies a region of interest based on predicted progress paths and scenarios, then concentrates high-precision sensing resources on this localized area while maintaining broader coverage at reduced intensity. This creates local quality enhancement where target vehicle detection precision is maximized in critical zones
4Reliability
If the vehicle control device performs collision avoidance control based on predicted progress path, then the accident prevention capability is improved, but the loss of time increases due to path prediction and scenario analysis processing
Solution Approach 1:
The system performs preliminary progress path prediction and scenario matching continuously in the background before collision risk arises. By maintaining ready-to-use predicted trajectories and pre-matched scenarios, the system eliminates real-time computation delays when actual collision avoidance decisions are required, enabling rapid response while maintaining comprehensive analysis
Solution Approach 2:
Once a high-risk scenario is identified through prediction and matching, the system skips detailed analysis of low-risk scenarios and directly proceeds to TTC calculation and collision avoidance control for the identified critical objects. This selective processing rushes through the decision pipeline for high-priority cases while maintaining thoroughness where needed
Data Source
AI summary
The present disclosure relates to a device for controlling a vehicle at an intersection. The device variably sets a region of interest of a vehicle sensor in accordance with a relationship between the progress paths of a subject vehicle and a target vehicle at an intersection, or controls the subject vehicle according to a collision avoidance control method in comparison with a predetermined scenario of a collision possibility of the subject vehicle and the target vehicle at the intersection according to a scenario, thereby preventing a collision at the intersection.