HD Map Misalignment Hotspot Detection for Autonomous Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional maps are inadequate for autonomous vehicles due to lack of accuracy, freshness, and cost-effectiveness, making it challenging to provide safe navigation.
Innovation Solution
The system generates high-definition maps by combining sensor data from vehicles to create a three-dimensional representation, identifying misaligned portions, and analyzing their accuracy, using techniques such as point cloud processing and machine learning-based filters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional survey teams use drivers with high resolution sensors to create maps, then map accuracy is improved, but the process becomes expensive and time consuming
Solution Approach 1:
The system enables vehicles to automatically generate and update HD maps using their own sensor data without requiring dedicated survey teams. Vehicles self-service the map creation process by contributing their sensor measurements to the collective map database, eliminating the need for manual survey operations while maintaining high accuracy standards
Solution Approach 2:
The system combines sensor data from multiple vehicles to generate a unified high-definition map. By merging measurements from numerous vehicles traveling along routes, the system achieves comprehensive coverage and high accuracy without requiring a single vehicle to complete exhaustive survey operations, significantly reducing time and cost
2Reliability
If survey fleets increase the number of cars to capture road updates, then map freshness is improved, but the cost increases
Solution Approach 1:
Each vehicle in the fleet contributes its sensor measurements to update the map, so every vehicle serves dual purposes: transportation and map maintenance. This eliminates the need to increase fleet size, as the existing vehicles automatically perform both functions simultaneously, maintaining map freshness without additional cost
Solution Approach 2:
Vehicles are designed to perform multiple functions: they serve as both transportation units and mobile survey platforms. The same sensors used for navigation and safety also capture data for map updates, allowing a single fleet to maintain current maps without requiring dedicated survey vehicles or increasing overall fleet size
3Ease of operation
If conventional maps are used for autonomous vehicle navigation, then navigation is provided, but the accuracy is insufficient for safe navigation
Solution Approach 1:
The system transitions from conventional two-dimensional map representations to three-dimensional high-definition maps that include vertical dimensions, surface normals, and detailed geometric information. This dimensional enhancement provides both navigation capability and precise location accuracy (10 cm or less) by capturing the full spatial complexity of the environment
Data Source
AI summary
A high-definition map system receives sensor data from vehicles travelling along routes and combines the data to generate a high definition map for use in driving vehicles, for example, for guiding autonomous vehicles. A pose graph is built from the collected data, each pose representing location and orientation of a vehicle. The pose graph is optimized to minimize constraints between poses. Points associated with surface are assigned a confidence measure determined using a measure of hardness/softness of the surface. A machine-learning-based result filter detects bad alignment results and prevents them from being entered in the subsequent global pose optimization. The alignment framework is parallelizable for execution using a parallel/distributed architecture. Alignment hot spots are detected for further verification and improvement. The system supports incremental updates, thereby allowing refinements of sub-graphs for incrementally improving the high-definition map for keeping it up to date.