Distributed Pose Graph Optimization for HD Map Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional maps for autonomous vehicles lack the accuracy and timeliness required for safe navigation, as they are often outdated and rely on expensive and time-consuming survey processes, while existing sensor systems face limitations in observing all environmental features, leading to potential navigation errors.
Innovation Solution
The development of high-definition (HD) maps that utilize pose graphs and distributed computing to generate and maintain accurate, up-to-date maps by combining data from multiple vehicles, allowing for real-time updates and precise location determination within autonomous vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional survey teams use drivers with specially outfitted cars and high resolution sensors to create maps, then map accuracy can be improved, but the process becomes expensive and time consuming
Solution Approach 1:
The patent divides the large-scale map creation task into smaller sub-tasks performed by multiple vehicles independently. Each vehicle captures and processes a portion of the map data, and these partial results are subsequently merged to form the complete high-definition map. This segmentation enables parallel processing, dramatically reducing the overall time required while maintaining accuracy through distributed data collection.
Solution Approach 2:
The patent enables ordinary vehicles to perform map creation and updating functions themselves without requiring specialized survey equipment or teams. Each vehicle uses its own sensors to capture environmental data, processes this data locally, and contributes to the collective map. This self-service approach eliminates the need for expensive specialized survey operations while continuously updating maps with fresh data from diverse sources.
2Reliability
If conventional maps are updated frequently to reflect road changes, then map freshness is improved, but the cost and complexity of maintaining maps increases
Solution Approach 1:
The patent implements continuous map updating by having vehicles constantly capture and transmit map data during normal operation. Instead of periodic survey missions, the system operates continuously as vehicles traverse the environment, constantly refreshing the map data. This continuous action ensures maps remain current without requiring complex scheduled update operations or specialized survey deployments.
Solution Approach 2:
The patent merges map creation and maintenance functions with ordinary vehicle operations. Multiple vehicles contribute their sensor data to a centralized system that continuously updates the high-definition map. This combining of routine vehicle usage with map maintenance eliminates the need for separate, complex update operations while ensuring continuous freshness through aggregated data from many sources.
3Measurement precision
If pose graph optimization is performed on the entire graph at once, then optimization accuracy is improved, but computational time and resources increase significantly
Solution Approach 1:
The patent segments the pose graph into multiple subgraphs that can be optimized independently and in parallel. Each subgraph represents a portion of the overall map and can be processed separately using available computational resources. After individual subgraph optimizations converge, the results are merged to achieve global optimization. This segmentation enables parallel computation, dramatically reducing total optimization time while maintaining accuracy through systematic merging of partial results.
Data Source
AI summary
According to an aspect of an embodiment, operations may comprise obtaining a pose graph that comprises a plurality of nodes. The operations may also comprise dividing the pose graph into a plurality of pose subgraphs, each pose subgraph comprising one or more respective pose subgraph interior nodes and one or more respective pose subgraph boundary nodes. The operations may also comprise generating one or more boundary subgraphs based on the plurality of pose subgraphs, each of the one or more boundary subgraphs comprising one or more respective boundary subgraph boundary nodes and comprising one or more respective boundary subgraph interior nodes. The operations may also comprise obtaining an optimized pose graph by performing a pose graph optimization. The pose graph optimization may comprise performing a pose subgraph optimization of the plurality of pose subgraphs and performing a boundary subgraph optimization of the plurality of boundary subgraphs.