Vehicle Localization Using Topological Map and Semantic Point Cloud
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous vehicle systems face challenges in accurately determining vehicle paths and avoiding objects in their environment, requiring expensive equipment and significant computational resources for perception and localization, especially when traveling on repeatedly used routes.
Innovation Solution
The development of a topological map using stereo images to generate nodes with six degree-of-freedom data and 3D image features, allowing a vehicle to determine its location and object positions using a monocular image without stereo cameras or lidar sensors, and updating this map with new data to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If stereo cameras or lidar sensors are used for perception and localization, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent creates a topological map that copies the essential spatial and visual characteristics of the environment (nodes with 3D image features and six-DoF data) to enable localization without requiring expensive stereo cameras or lidar sensors. The map serves as a lightweight representation that captures the necessary environmental information for accurate positioning.
Solution Approach 2:
The patent replaces expensive, complex sensing equipment (stereo cameras, lidar) with a simpler, more affordable alternative: a monocular image processing system that uses a pre-built topological map. This substitution achieves comparable localization accuracy using significantly cheaper and simpler components.
2Measurement precision
If stereo cameras or lidar sensors are used for perception and localization, then measurement precision is improved, but computational resources required increase
Solution Approach 1:
The patent performs computationally intensive processing in advance by building a topological map during a mapping phase. The map stores pre-processed 3D image features and six-DoF data at various nodes. During localization, the system only needs to compare current monocular images against this pre-built map, significantly reducing real-time computational requirements while maintaining high localization accuracy.
3Measurement precision
If a topological map with nodes is created using stereo images, then localization accuracy is improved, but device complexity increases
Solution Approach 1:
The patent divides the environment into discrete topological nodes, each containing 3D image features and six-DoF data. This segmentation allows the system to process and store environmental information in manageable, localized units rather than as a continuous complex model, reducing overall system complexity while maintaining localization accuracy.
Data Source
AI summary
A computer, including a processor and a memory, the memory including instructions to be executed by the processor to determine a plurality of topological nodes wherein each topological node includes a location in real-world coordinates and a three-dimensional point cloud image of the environment at the location of the topological node and process an image acquired by a sensor included in a vehicle using a variational auto-encoder neural network trained to output a semantic point cloud image, wherein the semantic point cloud image includes regions labeled by region type and region distance relative to the vehicle. The instructions include further instructions to determine a topological node closest to the vehicle and a six degree-of-freedom pose for the vehicle relative to the topological node closest to the vehicle based on the semantic point cloud data, determine a real-world six degree-of-freedom pose for the vehicle by combining the six degree-of-freedom for the vehicle relative to the topological node and the location in real-world coordinates of the topological node closest to the vehicle and determine a location and size of a three-dimensional object in the semantic point cloud image based on three-dimensional background subtraction using the three-dimensional point cloud image included in the topological node closest to the vehicle. The instructions include further instructions to improve the three-dimensional point cloud image included in the topological node based on the semantic point cloud image and the real-world six degree-of-freedom pose for the vehicle.