HD Map Reverse Rendering for Occlusion-Free Scene Visualization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in navigation due to the limitations of conventional maps, which lack the accuracy required for safe operation, as they often rely on sensor data that can be obstructed or unreliable, leading to potential errors in vehicle localization and decision-making.
Innovation Solution
A system generates high-definition map data using a 3D grid representation, performs reverse rendering to exclude obstructing objects, and dynamically updates images based on sensor data and map information, allowing for accurate localization and obstruction-free visualization of the vehicle's surroundings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional maps are used for autonomous vehicle navigation, then device complexity is reduced, but measurement precision and reliability deteriorate due to insufficient accuracy for safe navigation
Solution Approach 1:
The patent segments map data into multiple layers including geometry layers (road boundaries, lanes), attribute layers (speed limits, traffic rules), and object layers (signs, obstacles). This segmentation allows the system to use only necessary map layers for specific navigation tasks, achieving high precision location accuracy while managing data complexity through selective layer processing.
Solution Approach 2:
The patent transitions from conventional 2D map representations to 3D spatial models with elevation data and multi-layered map structures. This dimensional enhancement enables more precise location determination by incorporating vertical information and multiple map attributes, while the structured organization of 3D data maintains computational efficiency.
2Reliability
If sensor data is used for autonomous vehicle operation, then real-time information is obtained, but reliability deteriorates when sensors are obstructed or fail to detect objects early enough
Solution Approach 1:
The patent performs preliminary rendering of the environment based on map data before actual sensor detection occurs. By pre-identifying objects and their positions from high-definition maps, the system prepares expected environmental information in advance, allowing it to compensate for sensor obstructions and detect objects earlier than real-time sensing alone would permit.
Solution Approach 2:
The patent introduces map data as an intermediary between the vehicle and the physical environment. Instead of relying solely on direct sensor-to-object detection, the system uses pre-stored map information as a mediator to infer the presence and position of objects, particularly those obscured from sensor view, thereby enhancing detection reliability.
3Measurement precision
If high definition map data is used to improve localization accuracy, then measurement precision improves, but device complexity and data processing requirements increase
Solution Approach 1:
The patent applies local quality by processing and rendering only the specific portion of high-definition map data relevant to the vehicle's current location and field of view. Instead of processing entire map datasets, the system selectively renders map features within the visible environment, achieving high localization accuracy while minimizing computational complexity through localized data processing.
Solution Approach 2:
The patent implements partial action by rendering only essential map features needed for current navigation decisions rather than processing all available map data. The system selectively displays road boundaries, lanes, and relevant objects based on the vehicle's context, achieving sufficient localization accuracy without the computational burden of complete map processing.
4Loss of information
If all objects are rendered in the visualization, then completeness of information is improved, but clarity deteriorates due to obstruction by closer objects
Solution Approach 1:
The patent applies the inversion principle by rendering far objects first and then selectively excluding or transparentizing closer obstructing objects. Instead of the conventional approach of rendering all objects and hoping for the best visibility, the system inverts the process by prioritizing distant object rendering and actively removing visual obstructions, thereby maintaining both information completeness and visualization clarity.
Solution Approach 2:
The patent extracts and excludes obstructing objects from the rendered visualization when they block the view of more important far objects. By selectively removing these intermediate obstacles from the display while retaining their spatial information in the data structure, the system maintains complete environmental awareness while improving visual clarity for critical distant features.
Data Source
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.