Occupancy Grid Mapping With Segment-Based Probability Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The generation of maps of physical environments around sensors in autonomous and ADAS-equipped motor vehicles is complex due to the large amount of data processing required for occupancy grids, necessitating a simplified method to reduce calculation complexity.
Innovation Solution
A method using a polar coordinate system to measure object positions with distance and azimuth coordinates, identifying a segment representing probable values along one dimension, and calculating occupancy probabilities for cells using a probability density function, which simplifies the processing by reducing the number of calculations needed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional occupancy grid mapping methods are used to process sensor data, then accurate environment mapping is achieved, but calculation complexity and processing time increase significantly
Solution Approach 1:
The patent segments the occupancy probability calculation by identifying a subset of potentially occupied cells that are intersected by the measured object, rather than calculating probabilities for all grid cells. This segmentation reduces the number of calculations while maintaining accuracy for relevant areas.
Solution Approach 2:
The patent applies local quality by focusing computational resources on specific regions of the occupancy grid where objects are detected. By calculating occupancy probabilities only for cells intersected by the measured object and using segment-based probability density functions, the method concentrates accuracy where needed rather than uniformly across the entire grid.
2Measurement precision
If traditional occupancy grid mapping methods are used to process sensor data, then accurate environment mapping is achieved, but processing time increases
Solution Approach 1:
The patent segments the processing task by identifying only those grid cells that are potentially occupied based on object detection. By calculating occupancy probabilities only for this subset of cells rather than the entire grid, processing time is reduced while maintaining mapping accuracy for relevant regions.
Solution Approach 2:
The patent applies partial action by performing occupancy probability calculations only for potentially occupied cells intersected by measured objects, rather than calculating probabilities for all cells in the occupancy grid. This partial processing approach reduces computation time while sufficient for accurate environment mapping.
Data Source
AI summary
An illustrative example method of mapping a physical environment using an occupancy grid containing a set of cells associated with respective occupancy probabilities includes measuring a potential position of an object, using a sensor and identifying a segment representing a distribution interval of probable values associated with respective probability values relating to the measured potential position. The segment extends according to only one of two dimensions of the coordinate system of the sensor and penetrates a subset of potentially occupied cells. The method includes evaluating a probability of occupancy of each potentially occupied cell by determining the features of a segment portion included in the potentially occupied cell, and determining the probability of occupancy of the potentially occupied cell as a function of the determined segment portion using the probability density function.