Distributed Object Tracking System with Adaptive Thresholds

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current systems for tracking and analyzing dynamic processes with moving objects in limited spaces, such as football games, face challenges in real-time data processing and object identification, particularly in maintaining low latency and accuracy across multiple tracking units.

Innovation Solution

A distributed system with tracking units using sensors and cameras that determine spatial coordinates, assign local identification codes, and transmit data to a central processing unit for real-time analysis, incorporating adaptive threshold values and calibration processes to ensure accurate object tracking and situation analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple tracking units are used to monitor the entire playing field, then measurement precision and coverage are improved, but device complexity and data processing burden increase

Engineering Contradiction:
Improveobject location accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The playing field is divided into multiple detection zones, each monitored by a separate tracking unit. Each tracking unit independently processes objects within its zone, reducing the computational burden on any single unit while maintaining overall system precision through coordinated multi-unit operation.

Inventive Principle:
Principle #1Segmentation

2Productivity

If real-time analysis is performed at high frequency (12-25 times per second), then productivity and responsiveness are improved, but use of energy and processing power increase

Engineering Contradiction:
Improveanalysis speedVSAvoidprocessing energy consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

Tracking units perform preliminary detection and tracking of objects within their zones, pre-processing data before it reaches the evaluation unit. This preliminary action reduces the complexity of real-time analysis, enabling high-frequency processing while reducing the energy burden on the central evaluation unit.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system performs partial analysis at the tracking unit level (detection and tracking within local zones) and reserves full situation analysis for the central evaluation unit. This distributed partial action enables high-frequency operation without requiring all units to perform complete analysis at maximum frequency.

Inventive Principle:
Principle #16Partial or excessive action

3Ease of operation

If distributed tracking units assign local identification codes independently, then ease of operation and processing speed are improved, but loss of information and identification accuracy worsen

Engineering Contradiction:
Improvetracking management simplicityVSAvoidobject identification consistency
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The central evaluation unit receives tracking data from multiple units with different local identification codes and performs feedback-based correlation. It compares object characteristics, positions, and movement patterns across tracking units to establish consistent global identification, resolving identification inconsistencies while maintaining the operational simplicity of local independent coding.

Inventive Principle:
Principle #23Feedback

4Manufacturing precision

If adaptive threshold values are used for object assignment, then manufacturing precision and tracking accuracy are improved, but device complexity and computational requirements increase

Engineering Contradiction:
Improveobject assignment accuracyVSAvoidalgorithm complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The system uses adaptive threshold values that dynamically adjust based on object characteristics, detection conditions, and tracking history. This dynamic adaptation improves object assignment accuracy by optimizing thresholds for specific situations while the modular architecture keeps the increased computational complexity localized to individual tracking units rather than the entire system.

Inventive Principle:
Principle #15Dynamics

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The system provides continuous, low-latency analysis results comparable to human perception, ensuring accurate tracking and identification of objects with high precision, enabling real-time game analysis and subsequent data storage for further review.

Implementation Method 1

The determination itself can then be carried out using sonar waves or electromagnetic signals, for example based on the transit time determination or the Doppler effect

Methodology Applied
Scientific EffectTransit time determination: Time of Flight

Implementation Method 2

The determination itself can then be carried out using sonar waves or electromagnetic signals, for example based on the transit time determination or the Doppler effect

Methodology Applied
Scientific EffectDoppler effect: Doppler Effect

Data Source

PatentEP1864153B1Object-tracking and situation-analysis system
Publication Date: 2014.10.29 FLUYDS
  • EP1864153B1 patent drawingFigure 1
  • EP1864153B1 patent drawingFigure 2
  • EP1864153B1 patent drawingFigure 3

AI summary

The invention relates to a situation-analysis system for detecting and tracking moving objects in a limited space during a predefinable time period and for analysing object constellations and episodes over time, in addition to situations during the period, by means of a tracking unit. Said tracking unit comprises at least one sub-unit, which is located and configured in such a way that it can detect the limited space and the moving and fixed objects in said space and a central positional data-processing unit, which determines a global position for each actual object in the detected space and generates a global object identification and a corresponding time-dependent global positional data record. According to the invention, each tracking unit has a data transfer unit, connecting it to the central positional-data processing unit and permitting bi-directional communication, in such a way that the central positional-data processing unit can transmit data to each tracking unit individually or to several tracking units simultaneously via a backward channel.