Depth Camera Hand Tracking Assembly Workstation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing manual assembly workstations lack reliable hand tracking systems, which can lead to inefficiencies and errors in monitoring and controlling assembly processes.

Innovation Solution

A method utilizing a depth camera connected to a control unit to record actual and reference depth images, with a reference line defined on the work surface, allowing for accurate detection and tracking of assembly personnel's hands by identifying structures that connect to or intersect the line and possess a minimum area, thereby enabling reliable hand tracking.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a depth camera is used to record actual depth images in a detection area, then hand tracking capability is improved, but reliability of manual tracking remains insufficient due to lack of reference comparison

Engineering Contradiction:
Improvehand tracking accuracyVSAvoidmanual tracking reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The system performs preliminary action by recording a reference depth image of the detection area before assembly operations begin. This reference image serves as a baseline for comparing subsequent actual depth images, enabling reliable detection of hand movements and structures in the workspace.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements feedback by continuously comparing actual depth images against the reference depth image. This comparison provides feedback about changes in the detection area, allowing the control unit to reliably identify hands and arms of assembly personnel based on deviations from the reference state.

Inventive Principle:
Principle #23Feedback

2Measurement precision

If complex algorithms are used to identify hand structures in depth images, then tracking accuracy is improved, but device complexity increases

Engineering Contradiction:
Improvehand detection accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system extracts only the essential information needed for hand tracking by comparing depth images and identifying structures that connect to or intersect with the reference line. This extraction approach focuses computational effort on relevant features rather than analyzing entire images, reducing complexity while maintaining accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The reference line serves as an intermediary element that simplifies hand detection. Instead of directly analyzing complex hand shapes in depth images, the system uses the reference line as a mediator to identify hands through simpler geometric relationships (connections and intersections), thereby reducing algorithmic complexity.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Area of stationary object

If the detection area is fully monitored without reference comparison, then coverage is improved, but false detections increase due to lack of baseline

Engineering Contradiction:
Improvedetection area coverageVSAvoiddetection reliability
Core Design Contradiction:
Area of stationary objectVSReliability

Solution Approach 1:

The system performs preliminary action by recording a reference depth image of the entire detection area before assembly operations begin. This reference image captures the baseline state of the workspace, enabling reliable distinction between permanent structures and transient hand movements throughout the full detection area.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements feedback by continuously comparing actual depth images against the reference depth image across the entire detection area. This comparison provides feedback that distinguishes legitimate hand movements from other objects or changes in the workspace, maintaining high detection reliability throughout the full coverage area.

Inventive Principle:
Principle #23Feedback

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

This approach provides extremely reliable manual tracking at assembly workstations, simplifying evaluation and ensuring accurate detection of hand movements, reducing errors and improving process control.

Implementation Method 1

a depth camera (21), which is arranged above the work surface (12) and has a detection range (42) covering the work surface (12), and a control unit (17), wherein the method utilizes the depth camera (21) to record actual depth images in a detection area (42)

Methodology Applied
Scientific EffectTime of flight: Time of Flight

Data Source

PatentEP3467544B1Method for an assembly workstation, assembly workstation, device for carrying out said method, computer program and computer readable medium
Publication Date: 2021.12.08 ROBERT BOSCH GMBH
  • EP3467544B1 patent drawingFigure 1
  • EP3467544B1 patent drawingFigure 2~3
  • EP3467544B1 patent drawingFigure 4~5

AI summary

A method for hand tracking at an assembly workstation is disclosed. A depth camera captures images of a work surface within its detection range. The depth camera records depth images. A control unit uses the depth camera to capture a reference depth image, which is free of body parts. A reference line is also defined within the detection range. The depth images recorded by the depth camera are compared with the reference depth image, and if deviations occur, a structure is extracted. The control unit then examines whether the structure is an arm. This is the case if the structure touches the reference line and/or has a minimum size.