Road Geometry Estimation Using BEV and Transformer Embeddings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing road geometry tracking solutions face challenges with inadequate accuracy and high computational and memory requirements, making them less suitable for real-time implementations in autonomous vehicles.
Innovation Solution
A computer-implemented method and system that encodes images from vehicle cameras using machine-learning algorithms to extract image features, transforms these features into a Bird's Eye View representation, and decodes this representation using transformer-based algorithms to extract object embeddings, thereby improving road geometry estimation accuracy without significant increases in computational power or data storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional road geometry tracking methods are used, then computational resources and memory storage are consumed, but accuracy is insufficient
Solution Approach 1:
The patent segments the road geometry tracking problem into multiple discrete road objects (lane markers, road edges, barriers) and processes each object independently through detection, classification, and tracking stages. This segmentation allows the system to manage computational complexity by handling individual objects rather than the entire road scene as a single complex problem, while improving accuracy through focused object-specific processing
Solution Approach 2:
The patent transforms 2D image features from camera views into a 3D Bird's Eye View representation, adding a dimensional transformation to the processing pipeline. This dimensionality change enables more accurate road geometry estimation by providing a top-down perspective that better represents the actual spatial relationships of road objects, while the transformation is optimized to maintain computational efficiency
2Productivity
If traditional road geometry tracking methods are used, then computational resources and memory storage are consumed, but real-time performance is degraded
Solution Approach 1:
The patent performs preliminary detection and classification of road objects in the first stage, creating a structured representation that is then used in subsequent tracking stages. By pre-processing and organizing road object data before detailed tracking, the system reduces the computational burden during real-time operation, enabling faster processing while maintaining accuracy
Solution Approach 2:
The patent implements an efficient memory management strategy where processed road object data from previous time steps is selectively retained only when necessary for tracking continuity. The system discards redundant or outdated information while recovering and reusing essential geometric features, thereby reducing memory storage requirements and computational power consumption while maintaining real-time performance
Data Source
AI summary
The invention relates to a method for determining a representation of one or more road objects of a road for a vehicle traveling on the road. The method includes, for each time step out of a plurality of consecutive time steps, encoding images output from the one or more cameras of a vehicle using machine-learning algorithms trained to output image features of road objects depicted in an image provided as input to the machine-learning algorithms. The method further includes transforming a plurality of image features included in the encoded images to a Bird's Eye View (BEV) representation of the plurality of image features. The method also includes decoding the BEV representation to extract a set of object embeddings using transformer-based machine-learning algorithms. Further, the method includes outputting a position and class of each road object of the one or more road objects by decoding the extracted set of object embeddings.