Ground Profile Estimation Using Outlier Removal for Reflective Sensor Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately differentiating between the ground and objects due to inaccurate sensor data, particularly from reflective materials like glass and metal, which can lead to unsafe navigation scenarios.
Innovation Solution
A trajectory verification system that includes a secondary perception component to classify sensor data as either ground or object, using techniques such as M-estimator and machine-learned models, and corrects the ground profile by excluding sensor data points that contribute to inaccuracies, ensuring safe trajectory planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If sensor data from reflective materials (glass, metal) is used for ground profile estimation, then the quantity of sensor data is increased, but the measurement precision of ground profile deteriorates due to multi-path reflections causing aberrations
Solution Approach 1:
The patent extracts and removes inaccurate sensor data points from the ground profile estimation process. The system identifies sensor data points that are inconsistent with the estimated ground profile (outliers) and excludes them from subsequent estimations, thereby eliminating the harmful effect of reflective materials while maintaining the use of available sensor data.
Solution Approach 2:
The patent implements an iterative feedback mechanism where the ground profile estimation is continuously refined. The system estimates a ground profile, identifies inconsistent sensor data points, removes them, and re-estimates the ground profile using the cleaned data. This feedback loop progressively improves measurement precision by eliminating erroneous data from reflective materials.
2Productivity
If all sensor data points are used for ground profile estimation, then the productivity of data processing is improved, but the reliability of object detection deteriorates due to false positives and false negatives from inaccurate data
Solution Approach 1:
The patent extracts and removes inaccurate sensor data points that would cause false positives or false negatives in object detection. By identifying and excluding outlier data points from the ground profile estimation, the system maintains high processing productivity while improving detection reliability through cleaner data.
Solution Approach 2:
The patent changes the parameter set used for ground profile estimation by dynamically adjusting which sensor data points are included. The system modifies the data subset based on consistency with the current ground profile estimate, thereby improving reliability without significantly impacting processing productivity.
3Ease of operation
If sensor data with multi-path reflections is used, then the ease of operation is maintained (simple sensor data collection), but the difficulty of detecting and measuring ground profile increases due to aberrations from reflective materials
Solution Approach 1:
The patent implements a self-correcting system where the ground profile estimation process automatically identifies and corrects its own errors. The system uses the estimated ground profile itself to identify inconsistent sensor data points (outliers) and removes them, enabling the system to self-correct without external intervention while maintaining ease of operation.
Solution Approach 2:
The patent uses feedback from the ground profile estimation to improve detection accuracy. The system estimates the ground profile, uses this estimate to identify erroneous data points from reflective materials, removes them, and re-estimates. This feedback mechanism reduces the difficulty of detection while maintaining simple sensor data collection.
Data Source
AI summary
Inaccurate sensor data, such as by caused by reflections, may skew ground profile estimations that identify a profile or plane associated with a surface, such as a roadway. Correcting a ground profile that was based on inaccurate sensor data may include determining a location at which a slope of the ground profile meets or exceeds a maximum slope, determining a line associated with the location and having a slope of a magnitude equal to the maximum slope, and using the line to determine a subset of sensor data points that lies outside a region defined by the line. The subset of sensor data points that lies outside the region may be excluded from the set of sensor data points used to generate ground profile(s) and a new ground profile may be determined using the set of sensor data points, less the subset of sensor data points.