Ego-Vehicle Map Localization Distinguishing Temporary Static Objects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating environment models struggle to distinguish between parked vehicles and static objects, leading to localization errors in ego-vehicles as these objects are incorrectly mapped as static, causing inaccuracies when their state changes.
Innovation Solution
A method that acquires and updates environment maps by differentiating between permanent and temporary static objects using predefined occupancy probabilities for special areas, such as parking spots, and integrating sensor data to accurately classify objects as static or dynamic, thereby improving localization accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor data is used to detect static objects for map generation, then the map can be created with environmental objects, but parked vehicles are incorrectly mapped as permanent static objects leading to localization errors
Solution Approach 1:
The system dynamically updates the map by continuously monitoring object states and transitioning objects between temporary static and permanent static categories based on observed behavior over time, rather than statically classifying objects during initial map generation
Solution Approach 2:
The system uses feedback from continuous sensor data and object state monitoring to refine map accuracy, where localization results and detected object movements feed back into updating the classification of static objects in the map
2Ease of manufacture
If all detected static objects are mapped as permanent static objects, then the map generation process is simple, but localization errors occur when temporary static objects change position
Solution Approach 1:
The map structure is designed to be dynamic, allowing automatic reclassification of objects from temporary static to permanent static based on observed motion patterns over time, balancing initial simplicity with long-term accuracy
Solution Approach 2:
The system performs preliminary classification of objects as temporary static vs. permanent static during map generation, and preliminarily sets up the infrastructure for continuous monitoring and automatic updates
3Reliability
If the map is updated continuously to reflect object movements, then localization accuracy is maintained, but computational resources and processing time increase
Solution Approach 1:
The system performs partial updates by only reclassifying objects that exhibit movement patterns consistent with temporary static objects, rather than continuously reprocessing the entire map data set
Solution Approach 2:
The update frequency and depth are dynamically adjusted based on detected object behavior, increasing monitoring for objects showing movement patterns while reducing updates for confirmed permanent static objects
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The invention relates to a method for generating a map in an ego-vehicle for localization of the ego-vehicle comprising the following steps: - Acquiring (S1) a map with static objects of an environment of the ego-vehicle; - Acquiring (S2) general area information and information on special areas of the ego-vehicle's environment; - Acquiring (S3) a predefined occupancy probability (POcc1 - POccX); - Assigning (S4) respective static objects from the map to special areas; - Determining (S5) if the static objects are permanent static objects or temporary static objects based on the assignment of the special areas; - Updating (S6) the map with the information on temporary and permanent static objects; - Capturing (S7) the environment of the ego-vehicle by at least one sensor (2) located in the ego-vehicle; - Generating (S8) an environment representation based on the sensor data of the at least one sensor (2); - Determining (S9) static and dynamic objects in the environment of the ego-vehicle based on the environment representation; - Localizing (S10) the ego-vehicle in the updated map using the sensor data and/or environment representation.