3D Vision System for Robotic Teat Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing dairy milking systems face challenges in accurately positioning and maneuvering robotic arms due to the unpredictable movement of dairy livestock, varying teat positions, and interference from features like tails, leading to inefficiencies and inaccuracies in teat cup attachment and milk extraction processes.

Innovation Solution

A vision system equipped with a robotic arm, a 3D camera, and a processor that acquires 3D images, identifies teats, determines approach vectors, and sends instructions for teat cup attachment, allowing real-time adjustments to avoid legs and tails, and utilizes teat location information for precise teat cup placement, even for close, hidden, or offset teats.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a robotic arm is used for automated milking operations, then productivity is improved through automation, but reliability deteriorates due to unpredictable livestock movement and varying teat positions

Engineering Contradiction:

Improveautomation of milking operationsVSAvoidteat cup attachment accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system uses real-time 3D imaging and computer vision to dynamically track and compensate for livestock movement and teat position changes during milking operations, allowing the robotic arm to adapt its positioning continuously rather than relying on static pre-positioning

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The vision system provides continuous feedback on teat location and livestock movement, enabling the control system to adjust robotic arm positioning in real-time to maintain accurate teat cup attachment despite animal movement

Inventive Principle:

Principle #23Feedback

2Manufacturing precision

If the robotic arm moves closer to the teat for precise attachment, then manufacturing precision is improved, but the risk of interference from legs and tails increases

Engineering Contradiction:

Improveteat cup attachment precisionVSAvoidinterference from legs and tails

Core Design Contradiction:

Manufacturing precisionVSObject-affected harmful factors

Solution Approach 1:

The system transitions from 2D image processing to 3D depth map analysis, enabling the robotic arm to navigate around obstacles in three-dimensional space and select optimal approach vectors that avoid legs and tails while maintaining precise teat cup attachment capability

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Reliability

If real-time vision processing is implemented to track livestock movement, then reliability is improved, but use of energy increases due to continuous 3D imaging and processing

Engineering Contradiction:

Improvecompensation for livestock movementVSAvoidenergy consumption of vision system

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system captures a sequence of 3D images over time to predict livestock movement trends, allowing the robotic arm to proactively adjust its positioning based on predicted rather than reactive corrections, reducing the frequency of high-energy processing operations

Inventive Principle:

Principle #10Preliminary action

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 vision system enhances the accuracy and efficiency of robotic arm operations by enabling real-time detection and compensation for leg and teat movement, improving teat cup attachment success rates and reducing the need for hard coding, thus optimizing dairy milking processes.

Implementation Method 1

a three-dimensional (3D) camera disposed on the robotic arm... acquire a 3D image using the 3D camera, wherein each pixel of the 3D image is associated with a depth value

Methodology Applied

Scientific EffectTime of Flight: Time of Flight

Implementation Method 2

a laser configured to compute a distance based on an amount of time it takes an emitted light to return to the laser

Methodology Applied

Scientific EffectTime of Flight: Time of Flight

Data Source

PatentEP3685659B1Vision system with teat detection

Publication Date: 2023.06.07 TECHNOLOGIES HOLDINGS CORP

Figure 1
Figure 2
Figure 3

AI summary

A vision system (100) that includes a robotic arm (200) and a three-dimensional (3D) camera (136) operably coupled to a processor (102). The processor (102) is configured to acquire a 3D image (138) and identify a teat (203) of the dairy livestock (202) within the 3D image (138). The processor (102) is further configured to determine a distance between a portion of a robotic arm (200) and an approach vector for the teat, to compare the distance between the portion of the robotic arm (200) and the approach vector for the teat (203) and the attachment range threshold value, and to send instructions to the robotic arm (200) based on the comparison.