Collision Warning System Using Relative Acceleration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing collision warning and avoidance systems (CWAS) rely on assumptions of constant relative velocity between vehicles and obstacles, which can lead to inadequate warning times and increased accident rates, as they fail to accurately account for relative acceleration and provide sufficient alert times for drivers to react.
Innovation Solution
A method and system that processes multiple camera images to estimate time to collision (TTC) by determining relative acceleration between a vehicle and an obstacle, using image data to calculate instantaneous relative velocity and acceleration, allowing for more accurate prediction of collision courses without assuming constant velocity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If constant relative velocity assumption is used to determine TTC, then calculation complexity is reduced, but measurement precision of TTC deteriorates due to failure to account for relative acceleration
Solution Approach 1:
The system changes the parameters used for TTC calculation from assuming constant velocity to incorporating relative acceleration. By using multiple image frames to compute changes in scale and derive acceleration, the system adapts the calculation model to reflect actual motion dynamics, thereby improving measurement precision without requiring complex additional hardware
Solution Approach 2:
The patent replaces traditional mechanical sensing systems (accelerometers, radar) with a vision-based computational approach. By substituting optical imaging and image processing for mechanical measurement devices, the system achieves acceleration-aware TTC calculation while maintaining relative simplicity in implementation
2Measurement precision
If laser radar is used to determine distance and velocity, then measurement precision of TTC is improved, but device complexity and cost increase
Solution Approach 1:
The system creates a visual copy of the physical scene through camera imaging and processes this optical information to extract motion parameters. By copying scene information through multiple image frames and deriving acceleration from scale changes, the system replaces expensive laser radar hardware with a computational vision approach that achieves similar measurement precision
Solution Approach 2:
The patent substitutes optical imaging systems with mechanical sensing systems. Instead of using laser radar or accelerometers to measure distance and velocity, the system uses camera images and image processing to derive the same information, thereby reducing device complexity and cost while maintaining measurement accuracy
3Reliability
If additional warning time is provided to drivers, then accident prevention rate increases, but loss of time for system response increases
Solution Approach 1:
The system performs preliminary calculation of TTC using current acceleration data to predict future collision risk. By computing TTC based on present motion state and extrapolating forward in time, the system provides advance warning to drivers before collision actually occurs, enabling preventive action while maintaining accurate timing
Solution Approach 2:
The system dynamically adjusts TTC calculation to account for changing acceleration conditions. Rather than using fixed velocity assumptions, the system continuously updates acceleration estimates from sequential images and recalculates TTC accordingly, providing accurate real-time warnings that adapt to evolving traffic situations
Data Source
AI summary
A method of estimating a time to collision (TTC) of a vehicle with an object comprising: acquiring a plurality of images of the object; and determining a TTC from the images that is responsive to a relative velocity and relative acceleration between the vehicle and the object.