Vehicle Localization via Real-Time Height Map Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in localizing themselves within lane boundaries, especially during harsh environmental conditions such as snow or heavy rain, where lane boundaries become undetectable, making it difficult to determine vehicle location.
Innovation Solution
A system and method that utilize existing sensors on the vehicle to filter and process image data for static environmental elements, determine real-time height values, correlate them with defined height values from a map, and use this information to localize the vehicle and control its navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If lane boundaries are used for localization, then vehicle positioning accuracy is improved under normal conditions, but localization fails during harsh environmental conditions such as snow or heavy rain
Solution Approach 1:
The patent transitions from two-dimensional lane boundary detection to three-dimensional height map correlation. By utilizing the vertical dimension (height values of static elements) in addition to horizontal position, the system creates a more robust localization method that remains effective when traditional 2D lane markings are obscured by environmental conditions.
Solution Approach 2:
The system pre-processes sensor data to create height maps and identifies static elements in advance. By preparing height value data and correlating it with map data before localization is needed, the system ensures rapid and reliable positioning even when lane boundaries become undetectable during adverse weather.
2Loss of information
If GPS technology is used for navigation, then vehicle location determination is improved, but precision in determining exact lane position deteriorates
Solution Approach 1:
The patent combines GPS location data with height map correlation data to achieve both broad area localization and precise lane positioning. By merging these two complementary approaches, the system recovers the information lost in each individual method, achieving both global location awareness and precise lane-level positioning.
3Measurement precision
If additional sensors are added to improve localization capability, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system makes existing multi-functional sensors (cameras, lidars, radars) serve the dual purpose of both environmental perception for navigation and height map generation for localization. By extracting height information from already-deployed sensors, the patent avoids adding dedicated localization sensors, thereby maintaining system simplicity while achieving precise localization.
Data Source
AI summary
Systems and method are provided for localizing a vehicle. In one embodiment, a method includes: receiving, by a processor, sensor data from a sensor of the vehicle; filtering, by the processor, the sensor data for data associated with static elements of the environment; determining, by the processor, realtime height values associated with the static elements; correlating, by the processor, the realtime height values with defined height values associated with a map of the environment; localizing, by the processor, the vehicle on the map based on the correlated realtime height values and the defined height values; and controlling, by the processor, the vehicle based on the localizing.