Vehicle Self-Localization Quality Estimation Under Obstructed Visibility
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current self-localization methods for autonomous vehicles rely on GNSS systems, which are inadequate for precise positioning, and require up-to-date maps with visible features, making it challenging to estimate the quality of localization, especially in scenarios with obstructed visibility due to moving or static obstacles.
Innovation Solution
A method to estimate the quality of localization by determining the future visibility of landmarks using sensors, which involves detecting obstacles, estimating their movement trajectories, and recording projected positions in a map, allowing for the assessment of sensor visibility and calculating the covariance of landmark positions to determine localization quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GNSS systems are used for self-localization, then the system is simple to operate, but the measurement precision is inadequate for precise positioning
Solution Approach 1:
The patent combines multiple localization methods (GNSS, map matching, sensor-based landmark detection) into a unified self-localization system. The system integrates data from GPS/GLONASS satellites with map data and sensor observations to achieve precise positioning that exceeds what any single method could provide alone.
Solution Approach 2:
The system performs multiple functions simultaneously: it uses GNSS for coarse positioning, map matching for route-constrained localization, and sensor-based landmark detection for precise positioning. This multi-functional approach ensures accurate localization across diverse driving scenarios while maintaining system simplicity through integrated processing.
2Measurement precision
If up-to-date maps with visible features are required for self-localization, then the measurement precision improves, but the ease of operation deteriorates due to obstructed visibility
Solution Approach 1:
The system performs preliminary actions by predicting future visibility of landmarks before the vehicle reaches them. It uses the current trajectory and map data to determine which landmarks will become visible in the near future, allowing the system to prepare and assess localization quality in advance, even when landmarks are currently obscured by obstacles.
Solution Approach 2:
The system introduces an intermediary prediction mechanism that bridges the gap between current obscured visibility and future visible landmarks. By calculating which landmarks will be visible based on predicted vehicle position and trajectory, the system mediates between current sensor limitations and future localization opportunities.
3Reliability
If the system deactivates autonomous driving functions when localization criteria are not met, then the reliability improves, but the productivity decreases due to reduced operational time
Solution Approach 1:
The system performs preliminary assessment of localization quality by predicting future visibility of landmarks along the planned trajectory. It determines whether sufficient landmarks will be visible in the near future before committing to autonomous operation, allowing proactive safety decisions rather than reactive deactivation.
Solution Approach 2:
The system continuously monitors predicted localization quality and uses this feedback to make real-time decisions about autonomous driving function activation. The feedback loop compares predicted landmark visibility against required localization quality thresholds, dynamically adjusting system operation to maintain safety while maximizing productive operational time.
Data Source
AI summary
A method for estimating the quality of localization using sensor detection, wherein the vehicle detects dynamic objects on the road and in the direct surroundings of the road and estimates the dimensions of the objects. The movement of these objects in the near future is estimated. The outer casings of these objects are entered into a map of the surroundings. From the perspective of the sensors used to detect the features in the surroundings, the limitations of the fields of view and the predicted temporal development thereof resulting from the movement of the transportation vehicle and the predicted movements of the objects are entered into the map of the surroundings. The surrounding features that have been entered into the map of the surroundings and which may at visible in the near future are determined. An upper limit for a measure of the quality of localization is estimated.