Factor-Graph Lane Map Updates for Real-Time Vehicle Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles rely on pre-built maps that cannot be corrected in real-time, leading to errors in lane recognition and navigation, potentially causing vehicles to leave lanes unintentionally or fail to account for unanticipated road conditions.
Innovation Solution
A dynamically modifiable map using a factor graph with variable nodes and factor nodes that allow for real-time correction of lane line positions based on sensor data, enabling the vehicle to adjust its map while moving by updating the positions of variable nodes and their constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a pre-built static map is used for navigation, then the vehicle can perform navigation and localization operations, but the map cannot be corrected in real-time leading to errors in lane recognition and navigation
Solution Approach 1:
The patent transforms the static pre-built map into a dynamic map that can be modified in real-time. The system maintains a factor graph representation of the map where variable nodes can be updated with new sensor data during vehicle operation, allowing the map to adapt to actual road conditions while maintaining its structural integrity for navigation operations
Solution Approach 2:
The system implements feedback by continuously comparing sensor observations of lane lines with the stored map data. When discrepancies are detected, the system updates the factor graph with corrected information from sensors, creating a closed-loop system that self-corrects map errors in real-time based on actual road conditions
2Reliability
If the pre-built map is replaced with a corrected map after the vehicle stops, then map errors can be corrected, but the vehicle cannot correct deficiencies while moving and must rely on backup safety mechanisms
Solution Approach 1:
The system performs preliminary correction actions continuously during vehicle operation by maintaining a factor graph that can be updated with sensor data. Rather than waiting for the vehicle to stop, the system proactively corrects map errors in real-time by updating variable nodes in the factor graph with current sensor observations, eliminating the need for post-operation map replacement
3Reliability
If backup safety mechanisms are used to drive off camera centerline rather than pre-built map, then immediate safety can be maintained, but the deficiencies in the pre-built map are not corrected and may introduce issues with other autonomous vehicle operations
Solution Approach 1:
The patent merges the camera centerline information with the pre-built map data by integrating sensor observations into the factor graph. Instead of choosing between using the map or camera centerline, the system combines both sources by updating the factor graph with sensor data, allowing the vehicle to use corrected map information for navigation while maintaining safety through continuous sensor validation
Data Source
AI summary
Provided are systems and methods for controlling a vehicle based on a map that designed using a factor graph. Because the map is designed using a factor graph, positions of the road can be modified in real-time while operating the vehicle. In one example, the method may include storing a map which is associated with a factor graph of variable nodes representing a plurality of constraints that define positions of lane lines in a road and factor nodes between the variable nodes on the factor graph which define positioning constraints amongst the variable nodes, receiving an indication from the road using a sensor of a vehicle, updating positions of the variable nodes based on the indication and an estimated location of the vehicle within the map, and issue commands capable of controlling a steering operation of the vehicle based on the updated positions of the factor nodes.