Visual Trailer Angle Tracking via Reference Image Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Tow vehicle drivers face challenges in maneuvering vehicles attached to trailers due to confusion about steering angles, which can lead to incorrect turns and potential jack-knife situations.
Innovation Solution
A method for visual trailer tracking that involves receiving images from a rear camera on the tow vehicle, extracting visual features of the trailer, calculating the trailer angle relative to the tow vehicle, and storing these references for subsequent image matching and angle calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If visual feature matching is performed continuously to track trailer angle, then tracking accuracy is improved, but computational load and processing time increase
Solution Approach 1:
The system performs preliminary actions by storing reference images and their corresponding trailer angles during periods when the trailer is stationary or moving smoothly. These pre-captured reference data are then used for comparison during active maneuvering, eliminating the need for continuous real-time computation of absolute angles and reducing processing time during critical tracking moments.
Solution Approach 2:
The system creates copies of reference images and trailer angle data that are stored in memory for subsequent comparison. Instead of continuously calculating absolute trailer angles from current images, the system compares current images against stored reference copies, significantly reducing computational load while maintaining tracking accuracy.
2Reliability
If reference images are updated frequently to maintain tracking accuracy, then trailer angle estimation reliability is improved, but system complexity increases
Solution Approach 1:
The system implements feedback mechanisms where the measured trailer angle and image matching quality continuously inform whether reference images should be updated. When the trailer returns to a reference position or when significant changes occur, the system feedback-triggers an update of reference images, maintaining reliability without requiring constant updates and reducing unnecessary system complexity.
Solution Approach 2:
The reference image update mechanism is made dynamic rather than static. The system adapts the reference image update frequency based on trailer position, maneuvering state, and image matching quality. Reference images are updated when the trailer returns to reference positions or when degradation is detected, optimizing the balance between reliability and system complexity.
3Duration of action of moving object
If visual features are extracted from every received image, then tracking continuity is improved, but energy consumption increases
Solution Approach 1:
Instead of extracting visual features from every single received image, the system employs periodic action by selecting specific frames for feature extraction based on trailer maneuvering detection. Feature extraction is performed periodically at key moments (e.g., when maneuvering is detected or at regular intervals during stable conditions), reducing energy consumption while maintaining tracking continuity through interpolation and reference image comparison.
Data Source
AI summary
A method for visual trailer tracking of vehicle-trailer angle estimation, including: (a) camera receiving images, each most recently received image being identified as a current image, replacing a previously identified current image; (b) during a driving maneuver, extracting visual features from the current image, calculating a trailer angle from the current image, and storing the current image as a reference image, the trailer angle as a trailer angle reference and the extracted visual features as visual reference features; and (c) matching the visual features from the current image with the visual reference features. When a number of matches is below a threshold, repeating operation (c) to update the reference image, the trailer angle reference and the visual reference features. The method includes calculating a trailer angle between a trailer representation in the current image and a trailer representation in the reference image based in part upon the trailer angle reference.