Radar Disambiguating System for Ambiguous Detections

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Vehicle perception systems, particularly those using radar, often face ambiguity in detecting a single target, returning multiple candidate azimuth or elevation angles, which radar hardware cannot resolve, leading to inaccurate navigation in autonomous or semi-autonomous vehicles.

Innovation Solution

A disambiguating system comprising multiple modules that select the true detection based on proximity to predicted positions, known travel pathways, and data from other sensors, using techniques such as track gating, map matching, and machine learning to determine the most likely true detection among ambiguous radar detections.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Area of stationary object

If radar hardware is used for target detection, then detection coverage is improved, but measurement precision deteriorates due to ambiguous position detections

Engineering Contradiction:
Improvedetection coverageVSAvoidposition detection accuracy
Core Design Contradiction:
Area of stationary objectVSMeasurement precision

Solution Approach 1:

The patent introduces an intermediary processing system that receives ambiguous radar detections and resolves them by comparing with predicted target positions from tracking algorithms. This intermediary layer mediates between the radar hardware's limited precision and the navigation system's requirement for accurate position data, selecting the most likely true detection from multiple candidates based on proximity to predicted positions.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If multiple candidate detections are returned for a single target, then detection completeness is improved, but device complexity increases due to ambiguity resolution requirements

Engineering Contradiction:
Improvedetection completenessVSAvoidambiguity resolution complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by maintaining predicted target positions through tracking algorithms before new radar detections arrive. When ambiguous detections occur, the system immediately compares them against pre-computed predicted positions and selects the closest match, avoiding the need for complex real-time analysis of multiple candidates and simplifying the resolution process.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If radar ambiguity resolution is implemented, then measurement precision is improved, but loss of time occurs due to additional processing

Engineering Contradiction:
Improveposition detection accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent implements self-service by using the tracking system's own predicted positions to resolve radar ambiguities. The ambiguity resolution process serves itself by leveraging previously established target trajectories and predictions, eliminating the need for external reference systems or complex cross-validation with other sensors, thus minimizing additional processing time.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS10859673B2Method for disambiguating ambiguous detections in sensor fusion systems
Publication Date: 2020.12.08 GM GLOBAL TECHNOLOGY OPERATIONS LLC
  • US10859673B2 patent drawing
  • US10859673B2 patent drawing
  • US10859673B2 patent drawing

AI summary

A disambiguating system for disambiguating between ambiguous detections is provided. The system includes a plurality of modules, wherein each module is configured to disambiguate between ambiguous detections by selecting, as a true detection, one candidate detection in a set of ambiguous detections and wherein each module is configured to apply a different selection technique. The system includes: one or more modules configured to select as the true detection, the candidate detection whose associated position is closer to a position indicated by other data and one or more modules configured to select as the true detection, the candidate detection with the highest probability of being true based on other sensor data.