Occupancy Map Fusion with Adaptive Weighting for Sharper Boundaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing occupancy map fusion methods result in blurred boundaries between free and occupied space due to inconsistencies accumulating over time, particularly in dynamic and static conditions, making it difficult to accurately determine true boundaries.
Innovation Solution
A computer-implemented method that determines a fusion parameter representing potential dissimilarity between occupancy maps based on vehicle motion and sensor reliability, applying a fusion rule to control the influence of each map on the fused occupancy map, enhancing accuracy by adjusting weights and offsets per cell.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If occupancy maps are fused recursively over time to build up environmental information, then the completeness of environmental representation is improved, but the boundaries between free and occupied space become blurred
Solution Approach 1:
The patent applies dynamics by making the fusion parameter time-varying and adaptive. Instead of using a fixed fusion approach, the system dynamically adjusts the fusion parameter based on the temporal distance between current and historical occupancy maps, allowing the influence of historical data to decay over time. This resolves the contradiction by preventing blurring accumulation while maintaining environmental completeness.
Solution Approach 2:
The patent changes the parameter of fusion by introducing a time-dependent fusion parameter that varies with temporal distance. The fusion parameter is modified based on how much time has passed since the historical occupancy map was created, allowing the system to adaptively balance between utilizing historical information and maintaining boundary precision. This parameter change enables the system to reduce blurring while preserving environmental representation.
2Loss of information
If the influence of historical occupancy maps is increased to improve environmental representation, then the completeness of environmental information is improved, but inconsistencies accumulate over time causing blurring
Solution Approach 1:
The patent applies periodic action through the concept of temporal decay in the fusion parameter. Historical occupancy maps are fused with current maps, but their influence periodically decreases as time passes. This periodic reduction in influence prevents inconsistency accumulation while still utilizing historical information for environmental completeness, thereby resolving the contradiction between information completeness and consistency.
3Device complexity
If a fixed fusion approach is used to simplify the fusion process, then the device complexity is reduced, but the ability to adapt to different motion conditions is worsened
Solution Approach 1:
The patent applies dynamics by transitioning from a fixed fusion approach to a dynamic, adaptive fusion parameter that responds to vehicle motion conditions. The fusion parameter automatically adjusts based on temporal distance and motion state, providing adaptability to different scenarios without requiring complex manual configuration or multiple fusion strategies.
Solution Approach 2:
The patent applies self-service through the autonomous adaptation of the fusion parameter. The system automatically adjusts the fusion parameter based on temporal distance and vehicle motion without external intervention. This self-adjusting mechanism provides adaptability to different motion conditions while maintaining relatively simple fusion logic, resolving the contradiction between complexity and adaptability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for fusing a first occupancy map and a second occupancy map comprises: determining at least one fusion parameter representing a potential dissimilarity between the first occupancy map and the second occupancy map; determining a fused occupancy map representing free and occupied space around the vehicle, wherein the fused occupancy map is determined on the basis of the first occupancy map and the second occupancy map by means of a fusion rule, the fusion rule being adapted to control the influence of the first occupancy map and/or the second occupancy map on the fused occupancy map by means of the at least one fusion parameter.