Evidence Grid Occupancy Probability Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for combining multiple sensor measurements, such as those from radar systems, fail to effectively integrate data from different sensors and do not meaningfully update occupancy values in evidence grids, leading to incomplete representation of the sensed environment and potential conflicts in object detection.
Innovation Solution
A system that includes a processing device to calculate the probability of occupancy values in an evidence grid based on multiple detection signals from sensors, optimizing these values to maximize the match between the grid and the actual sensed environment, allowing for improved object detection and conflict resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple sensor measurements are combined using traditional evidence grid methods, then the representation of the sensed environment is updated, but the occupancy values do not meaningfully integrate data from different sensors, leading to conflicting information and indeterminate points
Solution Approach 1:
The patent transforms discrete occupancy values into continuous probability values ranging from 0 to 1, allowing for nuanced representation of object presence. This parameter transformation enables meaningful integration of multiple sensor measurements by computing probability distributions that reflect the cumulative evidence from different sensors, thereby resolving conflicts and eliminating indeterminate points while maintaining reliable occupancy representation.
2Productivity
If traditional evidence grid updating is used, then each measurement updates cells independently, but there is no meaningful combining of data from multiple measurements or different sensor types
Solution Approach 1:
The patent merges multiple sensor measurements by computing probability distributions that integrate evidence from different sensors and measurements. Instead of independently updating cells, the system combines probability information from multiple sources using statistical methods, thereby meaningfully integrating data while maintaining processing efficiency through probabilistic frameworks.
3Measurement precision
If multiple radar returns are combined to identify occupied or unoccupied volumes, then conflicting information from different sensors is revealed, but sensor errors and moving targets create indeterminate points that are neither occupied nor unoccupied
Solution Approach 1:
The patent implements dynamic probability updating where occupancy probabilities are continuously adjusted based on new sensor measurements. The system dynamically adapts to conflicting information by recalculating probability distributions that reflect the current state of evidence, allowing the occupancy determination to evolve as more data becomes available and resolving indeterminate points through cumulative probabilistic analysis.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A system (400) includes at least one sensor device (410) configured to transmit a first detection signal over a first spatial region and a second detection signal over a second spatial region. The second region has a first sub-region in common with the first region. The system further includes a processing device (420) configured to assign a first occupancy value to a first cell in an evidence grid. The first cell represents the first sub-region, and the first occupancy value characterizes whether an object has been detected by the first detection signal as being present in the first sub-region. The processing device (420) is further configured to calculate, based on the first and second detection signals, the probability that the first occupancy value accurately characterizes the presence of the object in the first sub-region, and generate a data representation of the first sub-region based on the probability calculation.