HD Map Visualization With Sensor Overlay for AV Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurate navigation due to limitations in conventional map data accuracy and sensor data reliability, particularly with obstructions and dynamic road conditions, which can lead to erroneous decisions and safety risks.
Innovation Solution
A system that combines sensor data from autonomous vehicles with high-definition map data to visualize and validate the vehicle's location, using motion compensation and semantic filtering to enhance localization accuracy and display dynamic updates, allowing for real-time correction of errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional maps are used for navigation, then the system is simple and easy to implement, but the accuracy is insufficient (over 100m error) for safe autonomous vehicle navigation
Solution Approach 1:
The patent combines multiple data sources including sensor data from the autonomous vehicle, high-definition map data, and localization algorithms to achieve accurate positioning. The system merges these diverse inputs to overcome the limitations of any single source, achieving the required 10cm or less accuracy for safe navigation.
Solution Approach 2:
The patent introduces an intermediary localization system that processes and reconciles data between GPS, sensor inputs, and map data. This intermediary layer performs motion compensation and data fusion to bridge the gap between conventional low-accuracy maps and the high-accuracy requirements of autonomous navigation.
2Reliability
If sensor data is used for real-time navigation decisions, then the system can adapt to dynamic conditions, but sensors may be obscured by obstructions and fail to observe everything
Solution Approach 1:
The patent implements feedback mechanisms where the system continuously compares sensor data with expected data from high-definition maps and localization algorithms. When discrepancies are detected due to obstructions or sensor failures, the system can identify and correct errors by referencing the more reliable map-based localization information.
Solution Approach 2:
The patent prepares for potential sensor failures by having pre-processed high-definition map data and multiple localization algorithms ready as backup. When sensor data becomes unreliable due to obstructions, the system can switch to or supplement with map-based positioning that was prepared in advance, cushioning against the reliability issues of sensor-only approaches.
3Measurement precision
If the system processes and displays combined sensor data and map data in real-time, then localization accuracy can be validated, but the computational load and processing time increase
Solution Approach 1:
The patent performs preliminary processing of map data and sensor data separately before combining them. High-definition map data is pre-processed and stored in an optimized format, and sensor data is pre-filtered and organized. This preliminary action reduces the computational burden during real-time combination, allowing accurate localization validation without excessive processing delays.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The autonomous vehicle generates an overlapped image by overlaying HD map data over sensor data and rendering the overlaid images. The visualization process is repeated as the vehicle drives along the route. The visualization may be displayed on a screen within the vehicle or at a remote device. The system performs reverse rendering of a scene based on map data from a selected point. For each line of sight originating at the selected point, the system identifies the farthest object in the map data. Accordingly, the system eliminates objects obstructing the view of the farthest objects in the HD map as viewed from the selected point. The system further allows filtering of objects using filtering criteria based on semantic labels. The system generates a view from the selected point such that 3D objects matching the filtering criteria are eliminated from the view.