Dynamic Obstacle Map Thresholds for Vehicle Sensor Noise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing obstacle detection methods in vehicle driver assistance systems, such as automated driving and parking assistants, face issues with accurately representing obstacles due to the faulty recognition of occupancy probabilities by 'Structure from Motion' (SFM) methods, leading to incomplete obstacle maps and noise suppression challenges.
Innovation Solution
A method that dynamically adjusts threshold values in the obstacle map based on the vehicle's trajectory and sensor detection properties, using a threshold pattern that varies by cell to account for the reliability of obstacle detection, ensuring accurate representation of obstacles while suppressing noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a constant threshold value is used to generate the obstacle map from the environment map, then the processing is simple, but obstacles at different distances and angles are not accurately detected due to varying sensor detection properties
Solution Approach 1:
The patent applies local quality by assigning different threshold values to different cells of the environment map based on their spatial characteristics (distance and angle relative to the vehicle). This allows each cell to be evaluated with an appropriate threshold that accounts for local sensor detection properties, thereby improving obstacle detection accuracy without requiring a completely complex system-wide redesign.
Solution Approach 2:
The threshold specification is made dynamic by adjusting threshold values according to the vehicle's trajectory and the specific cell characteristics. Instead of using a static constant threshold, the system adapts thresholds based on distance, angle, and movement history, enabling accurate detection across varying conditions while maintaining a relatively simple processing framework.
2Reliability
If the occupancy probability is accumulated from multiple SFM measurements, then more data is available for obstacle detection, but noise and false detections increase due to inconsistent recognition at different positions
Solution Approach 1:
The patent applies local quality by using position-dependent threshold values that are tailored to the specific characteristics of each cell (distance and angle). This allows the system to filter out noise and false detections locally by comparing occupancy probabilities against appropriate thresholds for each spatial location, rather than applying a uniform threshold that would either miss real obstacles or accept false ones.
Solution Approach 2:
The system uses feedback from the vehicle's trajectory and the spatial distribution of occupancy probabilities to dynamically adjust threshold values. By incorporating information about where the vehicle has been and how sensor detection properties vary with position, the system can feedback-adjust thresholds to suppress noise while maintaining sensitivity to real obstacles.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for providing an obstacle map, wherein the obstacle map comprises cells, which are associated with respective segments of the environment of the vehicle and each of which is assigned information as to whether the corresponding segment of the environment is occupied by an obstacle; wherein the method comprises: providing an environment map, wherein the environment map comprises cells, which are associated with respective segments of the environment of the vehicle and each of which is assigned an obstacle probability, which represents the probability that the corresponding segment of the environment is occupied by an obstacle; providing a threshold value specification; wherein the threshold value specification indicates different threshold values for cells of the environment map; wherein the threshold value specification is determined in accordance with the trajectory of the vehicle; determining the obstacle map on the basis of the environment map and in accordance with the threshold value specification.