Trailer Angle Detection Using Vision and Neural Network Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing trailer backup assist systems face challenges in accurately detecting the trailer angle due to variations in trailer hitch types, weather conditions, and environmental factors, leading to unreliable measurements.
Innovation Solution
A trailer angle identification system utilizing an imaging device and neural networks to process image data, steering angle data, and vehicle speed data, employing multiple neural networks to estimate and refine the trailer angle, ensuring accurate detection without additional sensors or markers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional trailer angle measurement systems are used, then the system structure is simple, but the measurement precision and reliability deteriorate due to variations in trailer hitch types, weather conditions, and environmental factors
Solution Approach 1:
The patent replaces traditional mechanical trailer angle sensors with a vision-based system using imaging devices and neural networks. The system captures images of the trailer and vehicle, processes them through convolutional neural networks to extract geometric features, and calculates the trailer angle without physical contact or mechanical components, thereby improving reliability across various hitch types and environmental conditions
Solution Approach 2:
The patent introduces neural networks as an intermediary between the raw image data and the trailer angle measurement. The neural networks process and interpret the visual information, extracting meaningful geometric relationships from images while being robust to variations in hitch types, weather, and environmental factors, thus improving measurement precision without requiring direct mechanical sensing
2Measurement precision
If multiple neural networks are used to process image data and steering angle data, then the measurement precision improves, but the device complexity increases
Solution Approach 1:
The patent divides the angle estimation task into multiple specialized neural networks: one network processes image data to estimate trailer angle, another processes steering angle and vehicle speed data to estimate angle change, and a third fuses these estimates. This segmentation allows each network to specialize in specific data types and estimation aspects, improving overall precision while maintaining manageable complexity through modular architecture
Solution Approach 2:
The patent combines multiple data sources (image data, steering angle data, vehicle speed data) and multiple neural network outputs through a fusion network. This merging of information streams and processing results creates a more robust and accurate trailer angle estimation by leveraging complementary information from different sensors and processing pathways
3Measurement precision
If additional sensors or markers are added to improve trailer angle detection, then the measurement precision improves, but the device complexity and cost increase
Solution Approach 1:
The patent enables the existing imaging device to serve multiple functions: capturing images for both trailer angle estimation and providing visual feedback to the driver during backup operations. The system extracts geometric features from normal operational images without requiring dedicated markers or additional sensors, allowing the same hardware to serve both measurement and user interface functions
Data Source
AI summary
A trailer angle identification system includes an imaging device configured to capture an image. A controller is configured to receive steering angle data corresponding to a steering angle of a vehicle, receive vehicle speed data corresponding to a vehicle speed, and estimate an angle of the trailer relative to the vehicle by processing the image, the steering angle data, and the vehicle speed data in at least one neural network.