HD Map Constrainedness for Vehicle Localization Confidence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional maps used by autonomous vehicles lack precision and accuracy, leading to challenges in safe navigation due to outdated data and limited sensor visibility, especially in environments with obscured inputs like corners or rolling hills, and require expensive and time-consuming survey processes to maintain up-to-date maps.
Innovation Solution
The use of high-definition (HD) maps that incorporate a measure of constrainedness, allowing vehicles to determine their location and navigate safely by leveraging 3D structures and sensor data, with the ability to update maps dynamically and efficiently, reducing storage needs through data compression and localization using Kalman filters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional maps are used for autonomous vehicle navigation, then the system is simpler and cheaper to implement, but the localization accuracy and navigation safety deteriorate due to outdated data and limited precision
Solution Approach 1:
The system pre-computes and stores constrainedness measures for all vehicle poses in the HD map before runtime use. This preliminary action allows the autonomous vehicle to quickly query pre-calculated localization confidence values without performing complex real-time computations, thereby achieving high measurement precision while avoiding real-time computational complexity
Solution Approach 2:
The patent adds a new dimension to traditional maps by incorporating constrainedness measures that quantify localization confidence. This transforms the map from a simple geometric representation into a multi-dimensional data structure that includes both spatial information and confidence metrics, enabling the vehicle to assess localization quality without increasing base system complexity
2Reliability
If survey teams use expensive high resolution sensors to create comprehensive maps, then map accuracy improves, but the time and cost to maintain up-to-date maps deteriorate due to the slow survey process
Solution Approach 1:
The system enables autonomous vehicles to self-update map data by collecting sensor measurements during normal operation and contributing them to the HD map. This self-service approach allows the map to automatically refresh itself using data from multiple vehicles, eliminating the need for expensive and time-consuming manual survey teams while maintaining high reliability and data freshness
Solution Approach 2:
The patent combines data from multiple autonomous vehicles into a single HD map system. By merging sensor data, constrainedness measures, and localization information from many vehicles operating in different locations and times, the system achieves comprehensive and up-to-date mapping without requiring dedicated survey operations, thereby reducing both time loss and costs
3Measurement precision
If HD maps store detailed 3D representations for high precision localization, then localization accuracy improves, but storage requirements and data processing complexity increase
Solution Approach 1:
The system extracts only the essential features needed for localization from complete 3D environmental data. Instead of storing full point clouds or detailed 3D models, the patent extracts key geometric features and pre-computes constrainedness measures from these features, thereby achieving high localization precision while dramatically reducing the quantity of stored data
Solution Approach 2:
The system performs preliminary processing of 3D data to pre-compute constrainedness measures and store only the essential localization information. By doing the heavy computational work beforehand and storing pre-processed results, the system reduces both storage requirements and real-time processing complexity while maintaining high measurement precision
4Adaptability or versatility
If sensors are used to detect road inputs in real-time, then the vehicle can respond to current conditions, but sensor visibility is limited by corners, rolling hills, and other obstacles
Solution Approach 1:
The system pre-stores information about road geometry, 3D structures, and environmental features in the HD map before the vehicle encounters them. This preliminary action allows the vehicle to access information about areas that are currently obscured from sensor view, compensating for limited real-time sensor detection capability and reducing information loss
Solution Approach 2:
The HD map acts as an intermediary between the vehicle's limited sensor capabilities and the complete environmental information. The map provides pre-captured data about road inputs, 3D structures, and geometric features that sensors cannot currently observe, thereby mediating the information gap caused by obscured views and improving overall adaptability
Data Source
AI summary
According to an aspect of an embodiment, operations may comprise accessing a set of vehicle poses of one or more vehicles; for each of the set of vehicle poses, accessing a high definition (HD) map of a geographical region surrounding the vehicle pose, with the HD map comprising a three-dimensional (3D) representation of the geographical region, determining a measure of constrainedness for the vehicle pose, with the measure of constrainedness representing a confidence for performing localization for the vehicle pose based on 3D structures surrounding the vehicle pose, and storing the measure of constrainedness for the vehicle pose; and for each of the geographical regions surrounding each of the set of vehicle poses, determining a measure of constrainedness for the geographical region based on measures of constrainedness of vehicle poses within the geographical region, and storing the measure of constrainedness for the geographical region.