3D Occupancy Grid Obstacle Detection for Self-Driving Cars
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle systems face challenges in accurately and quickly detecting objects around the vehicle, particularly in avoiding false negatives, which can lead to inadequate response times for potential collisions.
Innovation Solution
A method and system using an occupancy grid to predict the distance to the closest object by generating an occupancy grid with cells indicating the presence of objects, utilizing sensor data from cameras and LIDAR to determine the distance to the closest object, and employing machine learning algorithms for accurate object detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional sensor processing methods are used to detect objects around the vehicle, then the system can identify the presence of objects, but the detection speed and accuracy are insufficient, leading to inadequate response time for potential collisions
Solution Approach 1:
The patent divides the surrounding space into a grid of cells, with each cell independently processed to determine occupancy status. This segmentation allows parallel processing of multiple spatial regions simultaneously, improving both detection speed and accuracy while reducing the computational burden on any single processing unit
Solution Approach 2:
The patent transitions from traditional 2D sensor data representation to a 3D occupancy grid structure that incorporates spatial depth information. This dimensional enhancement allows the system to more accurately represent the three-dimensional space around the vehicle, improving object detection reliability while maintaining real-time processing capabilities
2Measurement precision
If the system processes all sensor data to ensure accurate object detection, then false negatives are reduced, but the processing time increases, reducing the speed of detection
Solution Approach 1:
By dividing the detection space into discrete grid cells, the system can process each cell independently and in parallel. This segmentation enables the system to maintain high detection precision through thorough analysis while achieving fast processing speeds through concurrent computation across multiple cells
Solution Approach 2:
The patent applies a level of processing that is sufficient to achieve high detection precision without being excessively complex. The occupancy grid method processes only the necessary spatial information at the appropriate level of detail, avoiding unnecessary computational overhead while maintaining high detection accuracy
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and an electronic device for determining a presence of an obstacle in a surrounding area of a self-driving car (SDC) are provided. The method comprises receiving sensor data representative of the surrounding area of the SDC in a form of 3D point cloud data; generating, by an MLA, based on the 3D point cloud data, a set of feature vectors representative of the surrounding area; generating, by the MLA, a grid representation of the surrounding area, each given cell of the grid representation including a predicted distance parameter indicative of a distance from the given cell to a closest cell with the obstacle; and using, by the electronic device, the distance parameter to determine presence of the obstacle in the surrounding area of the SDC.