Dynamic Occupancy Grid Fusion for Occluded Object Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous and semi-autonomous vehicle systems face challenges in accurately detecting and tracking objects, particularly smaller or occluded objects, due to limitations in sensor data fusion and object recognition techniques.
Innovation Solution
The proposed solution involves generating a dynamic occupancy grid based on sensor information from multiple sources, including cameras and radar, to identify dynamic cell clusters. This grid is then fused with low-level perception object detection results to generate an object track list, enhancing the detection of smaller or occluded objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If dynamic occupancy grid generation is implemented to improve object detection accuracy, then detection precision improves, but computational complexity increases
Solution Approach 1:
The environment is divided into discrete grid cells, with each cell independently processed to determine occupancy status. This segmentation allows parallel computation across cells, improving detection accuracy through detailed spatial analysis while managing computational complexity through divide-and-conquer processing
Solution Approach 2:
The patent transforms sensor data into a two-dimensional grid representation, adding a spatial dimension to object detection. This dimensional transformation enables more accurate detection of smaller and occluded objects by analyzing occupancy patterns across the grid, while the structured grid format facilitates efficient computation
2Measurement precision
If dynamic grid techniques are used to detect smaller or occluded objects, then detection capability improves, but data processing time increases
Solution Approach 1:
The system pre-generates the occupancy grid from sensor data before performing object detection. This preliminary grid creation organizes spatial information in advance, enabling faster detection of smaller and occluded objects during subsequent processing by leveraging the pre-structured occupancy patterns
Solution Approach 2:
The occupancy grid serves as an intermediary data structure between raw sensor input and object detection output. This intermediate representation captures environmental geometry and occupancy information, facilitating more efficient detection of difficult-to-detect objects while reducing direct processing of raw sensor data
3Reliability
If fusion of multiple data sources is implemented to improve object tracking, then tracking accuracy improves, but system complexity increases
Solution Approach 1:
The patent merges occupancy grid data with low-level perception object detection results into a unified object track list. This merging combines complementary information from different data sources, improving tracking accuracy by leveraging both geometric occupancy patterns and semantic object detection while integrating through a coordinated fusion process
Data Source
AI summary
Techniques are provided for utilizing a dynamic occupancy grid (DoG) for tracking objects proximate to an autonomous or semi-autonomous vehicle. An example method for generating an object track list in a vehicle includes obtaining sensor information from one or more sensors on the vehicle, determining a first set of object data based at least in part on the sensor information and an object recognition process, generating a dynamic grid based on an environment proximate to the vehicle based at least in part on the sensor information, determining a second set of object data based at least in part on the dynamic grid, and outputting the object track list based on a fusion of the first set of object data and the second set of object data.