3D Surfel Localization Using High-Stability Map Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle systems face limitations in representing three-dimensional environments, relying on either 2.5-dimensional maps that fail to capture multiple elevations or solely on current sensor data, which can be inefficient and unreliable due to changing environmental conditions.
Innovation Solution
The use of a three-dimensional surfel map that combines existing surfel data with real-time sensor data, allowing for accurate vehicle localization and decision-making by assigning uncertainty to both data sources, thereby leveraging previous knowledge and adapting to environmental changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a 2.5-dimensional map is used to represent the environment, then the system complexity is reduced, but the measurement precision of three-dimensional features is degraded
Solution Approach 1:
The patent transitions from 2.5-dimensional map representation to a full three-dimensional surfel-based representation. Each surfel is defined by three spatial coordinates (x, y, z) and three orientation parameters (pitch, yaw, roll), enabling accurate representation of multiple elevations and complex 3D structures like overpasses and trees at the same latitude/longitude location.
2Loss of time
If existing surfel map data is used for vehicle localization, then the processing time is reduced, but the reliability of localization is degraded due to environmental changes
Solution Approach 1:
The system dynamically fuses historical surfel map data with real-time sensor measurements to adapt to environmental changes. The fusion process weights and combines predictions from the static map with current sensor observations, allowing the system to maintain accurate localization even when the environment has changed since the map was created.
3Reliability
If only current sensor data is used for vehicle localization, then the reliability is improved, but the processing time and computational power increase
Solution Approach 1:
The system performs preliminary processing by maintaining pre-built surfel maps from historical data, which contain pre-extracted environmental features and structures. During localization, the system only needs to fuse these pre-processed maps with current sensor data, significantly reducing the computational burden compared to processing raw sensor data from scratch.
4Measurement precision
If a 3D mesh representation is used to integrate sensor data, then the measurement precision is improved, but the device complexity and processing requirements increase significantly
Solution Approach 1:
The patent segments the environment into discrete surfel units, where each surfel represents a localized surface element with defined spatial and orientational properties. This segmentation allows independent processing and updating of individual surfels, avoiding the need to reprocess entire 3D meshes when integrating new sensor data.
Solution Approach 2:
The system changes the representation parameters from continuous 3D mesh vertices and faces to discrete surfel elements characterized by position (x, y, z) and orientation (pitch, yaw, roll) parameters. This parameter transformation simplifies the integration of new sensor data, as each surfel can be independently updated or added without affecting the entire environment model.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using surfels for vehicle localization. One of the methods includes obtaining surfel data comprising a plurality of surfels, wherein each surfel corresponds to a respective different location in an environment, and each surfel has associated data that comprises a stability measure, wherein the stability measure characterizes a permanence of a surface represented by the surfel; obtaining sensor data for a plurality of locations in the environment, the sensor data having been captured by one or more sensors of a first vehicle; determining a plurality of high-stability surfels from the plurality of surfels in the surfel data; and determining a location in the environment of the first vehicle using the plurality of selected high-stability surfels and the sensor data.