Vehicle Environment Mapping with Radar Concealment Filtering

Solution Overview

Problem

Existing object tracking systems for autonomous vehicles suffer from false detections due to concealed objects closer to the radar sensor, leading to inefficient map creation and increased false positives from reflections or ground clutter.

Innovation Solution

A method that maps the environment by using only unconcealed objects detected by the radar sensor, creating a polygon with equally spaced circle segments and removing 'spikes' based on Euclidean distance thresholds, and incorporating virtual objects to enhance detection certainty, while tracking stationary objects over time to refine the map.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If all detected objects are included in the map, then the map coverage is complete, but false detections from concealed objects increase map complexity and reduce reliability

Engineering Contradiction:

Improvedetection reliabilityVSAvoidmap complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts and removes false detections from the map by identifying concealed objects that should not be included. The system distinguishes between valid detections and false detections caused by concealment, extracting only the reliable detections for map creation.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent performs preliminary identification of concealed objects before final map creation. By detecting which objects are concealed by closer objects and removing them in advance, the system prevents false detections from contaminating the final map.

Inventive Principle:

Principle #10Preliminary action

2Quantity of substance

If concealed objects are included in detections, then detection coverage is maximized, but false positives from reflections and ground clutter increase

Engineering Contradiction:

Improvedetection quantityVSAvoiddetection accuracy

Core Design Contradiction:

Quantity of substanceVSReliability

Solution Approach 1:

The patent converts the harmful effect of concealed objects generating false detections into a beneficial filtering process. By identifying the presence of concealing objects, the system systematically removes false detections from reflections and ground clutter, improving overall detection accuracy.

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

Solution Approach 2:

The system performs preliminary identification of concealed objects before final map creation. By detecting which objects are concealed by closer objects and removing them in advance, the system prevents false detections from contaminating the final map.

Inventive Principle:

Principle #10Preliminary action

3Loss of information

If map creation includes all detected objects, then environmental coverage is complete, but processing time and computational effort increase

Engineering Contradiction:

Improveenvironmental information completenessVSAvoidmap creation time

Core Design Contradiction:

Loss of informationVSLoss of time

Solution Approach 1:

The patent extracts and removes false detections from the map by identifying concealed objects that should not be included. The system distinguishes between valid detections and false detections caused by concealment, extracting only the reliable detections for map creation.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent performs preliminary identification of concealed objects before final map creation. By detecting which objects are concealed by closer objects and removing them in advance, the system prevents false detections from contaminating the final map.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11209539B2Method for mapping the environment of motor vehicles

Publication Date: 2021.12.28 FORD GLOBAL TECH LLC

AI summary

A computer includes a processor and a memory, the memory storing instructions executable by the processor to detect a plurality of objects around a vehicle and develop a map of an environment around the vehicle based on a polygon having sides extending between the plurality of objects, wherein the vehicle is located within an interior of the polygon.