Ghost Object Detection via Graph Context Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for detecting objects in a vehicle's environment fail to effectively distinguish between true and ghost objects, leading to potential dangerous situations in automated driving due to false-positive sensor measurements.

Innovation Solution

A method that utilizes map context and social context information to generate a graph structure representing the traffic situation, allowing for the classification of ghost objects by analyzing anomalies and patterns within this structure, incorporating data from multiple sensors and learning-based classification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If sensor measurements are used to detect objects in the vehicle's environment, then object detection capability is improved, but false-positive detections of ghost objects occur

Engineering Contradiction:
Improveobject detection accuracyVSAvoidfalse-positive rate
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

A graph structure serves as an intermediary representation that integrates sensor measurements with map context and social context information. This graph models spatial relationships and object interactions, enabling the system to cross-validate detections against expected environmental patterns before classifying objects as true or ghost objects.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system implements feedback loops where detection results are continuously validated against the graph structure representing map context and social context. Ghost objects are identified when sensor measurements contradict the established contextual model, and this feedback refines the classification process to reduce false positives.

Inventive Principle:
Principle #23Feedback

2Measurement precision

If contextual information from map data and multiple sensors is integrated, then ghost object detection accuracy is improved, but system complexity increases

Engineering Contradiction:
Improveghost object classification accuracyVSAvoidsystem architecture complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system segments contextual information into distinct modules: map context information from digital road maps, social context information from multiple sensors, and their integration in a graph structure. This segmentation allows each module to be processed independently and combined systematically, managing complexity through modular architecture.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The graph structure serves multiple functions simultaneously: it stores map context information, integrates social context from various sensors, models spatial relationships, and provides a framework for detecting anomalies. This multi-functionality reduces overall system complexity by consolidating multiple processing tasks into a single unified structure.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP4469943B1Method for detecting ghost objects in sensor measurements
Publication Date: 2025.12.03 MERCEDES BENZ GROUP AG
  • EP4469943B1 patent drawingFigure 1
  • EP4469943B1 patent drawingFigure 2
  • EP4469943B1 patent drawingFigure 3

AI summary

The invention relates to a method for detecting ghost objects in sensor measurements of an environment of a vehicle. According to the invention, map context information is obtained from a digital road map. Objects (O) with associated attributes are recognized in the environment of the vehicle, and social context information about the objects (O) in relation to each other is generated. All available data regarding a traffic situation involving the vehicle and the objects (O) is stored in a graph structure (GS), which comprises nodes (K1 to Km) and edges (E1 to En). Relational information (RI) is represented in the graph structure (GS) by means of the edges (E1 to En). In light of the map context information and the social context information, anomalies and patterns are searched for in features of the graph structure (GS), and ghost objects are classified on the basis of recognized anomalies and patterns.