Track Position Mapping Using Visual Features Between Timing Loops
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining a vehicle's position on a track, such as GNSS and timing loops, are not accurate enough, especially when the vehicle is traveling between loops or experiencing wheel slip or spin.
Innovation Solution
An apparatus and method using a camera-mounted vehicle to capture images and perform feature recognition, adjusting the vehicle's position on a track map based on recognized features' consistency across multiple laps, with indicators determining suitability for positioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GNSS is used to determine vehicle position, then the position can be determined continuously, but the accuracy is only within a few metres which is not sufficient for precise track positioning
Solution Approach 1:
The patent combines multiple positioning methods (GNSS, timing loops, and visual feature recognition) into a unified system. The visual recognition system processes images from onboard cameras to identify track features and calculate vehicle position, while timing loops provide reference points and GNSS provides continuous tracking data. This merging of methods allows the system to achieve high accuracy through visual feature matching while maintaining continuous tracking reliability through multiple data sources.
Solution Approach 2:
The patent introduces visual feature recognition as an intermediary between the vehicle and the positioning system. Instead of directly using GNSS coordinates or timing loop data, the system uses image processing to identify track features (curbs, barriers, signage) and matches them against a stored map. This intermediary layer enables precise position determination by comparing visual features with the track map, achieving accuracy beyond what GNSS alone can provide.
2Measurement precision
If timing loops are integrated into the track, then accurate position can be determined at loop positions, but no accurate information is available when the vehicle is travelling between timing loops
Solution Approach 1:
The patent divides the track into multiple visual feature segments rather than relying on a few discrete timing loop points. By identifying and tracking multiple visual features (curbs, barriers, signage, painted markings) distributed throughout the track, the system creates continuous position information between timing loops. Each visual feature acts as a segmentation point that provides positioning data, eliminating the information gaps that exist between traditional timing loop locations.
Solution Approach 2:
The patent performs preliminary actions by pre-mapping the track and storing visual feature data in a database before the vehicle arrives. The system captures images during practice sessions or previous laps, identifies visual features, and creates a reference map. When the vehicle is racing, this pre-prepared visual map enables immediate position determination without waiting for timing loop passages, eliminating information gaps through advance preparation.
3Device complexity
If wheel speed is used to determine position, then the system is simple to implement, but accuracy is compromised when the vehicle slides or skids or when wheel spin occurs
Solution Approach 1:
The patent replaces the mechanical wheel-speed-based positioning system with an optical vision-based system. Instead of relying on mechanical wheel rotation to calculate distance traveled, the system uses onboard cameras to capture images of visual features and processes these images to determine position. This substitution eliminates the fundamental problem of wheel slip affecting accuracy, as the visual recognition system independently measures position based on track features rather than wheel rotation.
Solution Approach 2:
The patent introduces visual feature recognition as an intermediary between the vehicle's motion and the positioning calculation. Rather than directly using wheel speed data, the system uses image processing to identify track features and calculate position based on the vehicle's relationship to these fixed visual landmarks. This intermediary visual measurement system provides accurate positioning independent of wheel slip or spin conditions.
4Measurement precision
If visual feature recognition is used to adjust position on map, then more accurate continuous tracking is achieved, but the system complexity increases due to image processing and feature matching requirements
Solution Approach 1:
The patent performs preliminary action by pre-capturing images of the track during practice sessions or previous laps and storing them in a database with identified visual features. This pre-processing creates a reference map that contains information about visual feature locations and appearances. During the actual race or time trial, the system only needs to compare current images against this pre-prepared reference, significantly reducing the computational complexity of real-time image processing while maintaining high positioning accuracy.
Solution Approach 2:
The patent applies partial action by selectively processing only certain visual features in the images rather than analyzing every pixel or feature. The system identifies and tracks specific key visual features (such as distinctive curbs, barriers, or signage) that provide sufficient positioning information. This selective approach reduces computational complexity compared to full-image processing while maintaining adequate positioning accuracy for the application.
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
The present technique provides an apparatus for determining a position of a vehicle on a track, the apparatus comprising: a storage medium operable to store a map of the track comprising a representation of one or more features of the track and to store a position of the vehicle on the map; input circuitry operable to receive, from a camera attached to the vehicle, a plurality of images captured by the camera as the vehicle travels along the track; and processor circuitry operable, for at least one of the plurality of captured images: to perform feature recognition on the image, to assign an indicator to at least one recognised feature in the image indicative of the suitability of the at least one recognised feature for adjusting the position of the vehicle on the map based on the recognised feature, and for each recognised feature in the image with an indicator which indicates that the recognised feature is suitable for adjusting the position of the vehicle on the map based on the recognised feature: when the recognised feature is already represented in the map of the track, to adjust the position of the vehicle on the map based on the position of the representation of the recognised feature in the map, and when the recognised feature is not already represented in the map of the track, to add a representation of the recognised feature to the map.