Autonomous Vision System Slippery Road Detection via Vehicle Motion Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vision systems for motor vehicles face challenges in reliably estimating slippery road conditions ahead, especially under poor light conditions, requiring expensive sensors and struggling to accurately detect road surface conditions without direct visual confirmation.
Innovation Solution
A cost-effective vision system incorporating a vehicle motion estimator and trajectory predictor, which estimates the motion of other vehicles and predicts a normal trajectory based on captured images and auxiliary information, allowing for the detection of slippery road conditions by comparing estimated motion with predicted trajectories, thereby indirectly determining side-slip angles without specialized sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If polarization filters or infrared sensors are used to detect water and ice puddles on the road ahead, then the detection capability under poor light conditions is improved, but the system cost increases significantly
Solution Approach 1:
The patent uses a regular camera to capture images of other vehicles and creates a virtual model of their motion behavior. This virtual model serves as a proxy to indirectly infer road conditions, replacing the need for expensive specialized sensors like polarization filters or infrared sensors while maintaining detection capability
Solution Approach 2:
The patent introduces other vehicles as intermediary objects to detect road conditions. Instead of directly sensing the road surface, the system observes the motion of other vehicles which are affected by road conditions, using them as mediators to infer the underlying road state
2Reliability
If direct optical sensing of the road surface is used to estimate slippery conditions, then the measurement capability is improved, but the system becomes more complex and less reliable under poor light conditions
Solution Approach 1:
Instead of directly observing the road surface to detect slippery conditions, the patent inverts the approach by observing other vehicles and inferring road conditions from their motion patterns. This indirect method proves more reliable under poor light conditions where direct road surface observation fails
3Measurement precision
If wheel velocity sensors or acoustic sensors are used to estimate friction coefficient at the vehicle's current position, then the measurement capability is improved, but the system cannot detect slippery conditions ahead of the vehicle
Solution Approach 1:
The patent performs preliminary detection of slippery conditions ahead of the vehicle by analyzing the motion of other vehicles before the ego vehicle reaches those road sections. This allows preventive actions to be taken before the vehicle encounters the hazardous conditions
Solution Approach 2:
The regular camera system serves multiple functions: capturing images for standard vision tasks and simultaneously analyzing other vehicles' motion to detect road conditions ahead, eliminating the need for separate specialized sensors
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A vision system (1) for a motor vehicle (2) comprises an imaging apparatus (3) adapted to capture images (5) from a surrounding (6) of the motor vehicle (2), a data processing device (7) providing a vehicle detector (14) adapted to detect another motor vehicle (13) in the surrounding (6) of the ego vehicle (2) by processing the images (5) captured by the imaging apparatus (3), and a slip detector (8) adapted to detect a slippery condition of a road (29) ahead of the ego motor vehicle (2). The vision system (1) comprises a vehicle motion estimator (9) and a trajectory predictor (10), wherein the vehicle motion estimator (9) is adapted to estimate the motion of the other motor vehicle (13) based on the output (30) of the vehicle detector (14), and the trajectory predictor (10) is adapted to predict a normal condition trajectory (12) of the other motor vehicle (13) based on the output (30) of the vehicle detector (14) and auxiliary information (17), wherein the slip detector (8) detects a slippery condition of the road (29) on the basis of a comparison between the output (11) of the vehicle motion estimator (9) and the output (12) of the trajectory predictor (10).