Location-Based Object Perception for Low-Lying Obstacle Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional perception systems struggle to accurately detect and localize low-lying or near-ground obstacles, such as wheel stops, curbs, and ground locks, due to weak sensor returns, which hampers the navigation and decision-making capabilities of autonomous or semi-autonomous machines.
Innovation Solution
Systems leverage location-based knowledge by determining regions of interest, such as parking spaces, and using geometric information to enhance the detection and tracking of these obstacles by analyzing sensor returns within defined height thresholds and correlation metrics, thereby improving perception accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional sensor-based perception techniques are used, then the system can detect most objects in the environment, but low-lying or near-ground obstacles cannot be accurately detected or localized due to weak sensor returns
Solution Approach 1:
The system performs preliminary action by determining target regions (such as parking spaces) and target areas within those regions before attempting to detect obstacles. By pre-defining where obstacles are likely to be located based on geometric information and predictable locations, the system can then focus its detection efforts on these specific areas, improving the reliability of detecting low-lying obstacles that would otherwise have weak sensor returns.
2Measurement precision
If the system focuses on detecting low-lying obstacles in specific target areas, then perception accuracy for these obstacles improves, but the system complexity increases due to additional processing steps
Solution Approach 1:
The system applies segmentation by dividing the environment into target regions (such as parking spaces) and further into target areas within those regions where specific obstacles are likely to be found. This segmentation allows the system to focus computational resources on specific areas rather than processing the entire environment, improving perception accuracy for low-lying obstacles while managing system complexity through localized analysis.
Data Source
AI summary
In various examples, certain objects commonly found in predictable locations of an environment may be more reliably perceived by leveraging known information related to the locations and/or the objects themselves. For instance, the disclosed systems and methods may determine locations of target areas within a coordinate system associated with a target region in an environment, and use the target areas to detect and track certain objects that may be otherwise difficult to perceive. As an example, a target region may be a parking space for a machine and the coordinate system may indicate target areas corresponding to wheel stops, curbs, ground locks, or other objects commonly associated with parking spaces. The systems may sample various points representing sensor returns to determine whether a target object is located in a target area, as well as to track the target object, in some instances.