Onboard Road Perception Using Associated-Region Lane Attributes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The accuracy of onboard perception models in vehicles for perceiving road environments is not high, which affects driving safety and efficiency.
Innovation Solution
The method involves acquiring associated-region lane attributes and information to be detected using onboard sensors, processing them with an onboard perception model to obtain road perception information, leveraging deep learning algorithms like convolutional neural networks and attention networks to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If onboard perception models are used for road environment perception, then driving automation is enabled, but perception accuracy is insufficient
Solution Approach 1:
The patent combines lane attribute information from associated regions with target region sensor data through data fusion. The perception model integrates lane line position, orientation, and curvature attributes from multiple sources to enhance overall perception accuracy and reliability for autonomous driving decisions.
Solution Approach 2:
The system performs preliminary extraction of lane attributes from associated regions before processing the target region. By pre-processing and storing lane attribute information from surrounding areas, the system prepares reference data that accelerates and improves the accuracy of real-time target region perception.
2Measurement precision
If lane attribute information from associated regions is integrated, then perception accuracy improves, but data processing complexity increases
Solution Approach 1:
The patent segments the road environment into target regions and associated regions, allowing independent processing of lane attributes for each segment. This segmentation enables modular data handling where lane attributes are extracted and processed separately for different spatial zones before being integrated, reducing overall processing complexity.
Solution Approach 2:
The perception model acts as an intermediary that receives lane attribute information from associated regions and sensor data from the target region, then processes and fuses these data streams. This intermediary processing layer manages the complexity by providing a structured interface between multiple data sources and the final perception output.
Data Source
AI summary
A method for perceiving a road environment, a vehicle control method, a training method, an electronic device, an autonomous driving vehicle, and a storage medium, which relate to fields of artificial intelligence technology, computer vision, deep learning and large model technologies, and may be applied to scenarios such as autonomous driving and unmanned driving. The method for perceiving a road environment includes: acquiring an associated-region lane attribute and an information to be detected, where the information to be detected is collected by an onboard sensor and represents a target region where a vehicle is traveling, the associated-region lane attribute corresponds to an associated region, and the associated region and the target region meet a predetermined similarity condition; and processing the associated-region lane attribute and the information to be detected by using an onboard perception model to obtain a road perception information of the target region.