SlideShare a Scribd company logo
1 of 25
National Taiwan University
Graduate Institute of Electrical Engineering
Modular ROS Based Autonomous Mobile
Industrial Robot System for Automated Intelligent
Manufacturing Applications
Ren C. Luo, Shang Lun Lee, Yu Cheng Wen, and Chin Hao Hsu
Department of Electrical Engineering, National Taiwan University, Taipei, Taiwan
Presenter: Shang Lun Lee
2020 IEEE/ASME International Conference on Advanced Intelligent Mechatronics
(AIM). Boston, USA. July 9, 2020
1
National Taiwan University
Graduate Institute of Electrical Engineering
 Background
 Objective
 Our Robotic Platform
 System Architecture
 SLAM & Navigation
 Manipulation
 Object Detection
 Grasp Pose Estimation
 Experiment
 Demonstration
 Conclusion
Outline
2
National Taiwan University
Graduate Institute of Electrical Engineering
Background (1/2)
3
 Mobile manipulator is a growing robotic machine such as service robot and AMR+cobot
 Small platform with 2~5kg arm payload v.s. large platform with 5~kg arm payload
 With mobility + dexterity + perception, they can do various tasks
-------- Service robot --------
Pr2 Pepper RB-1 Aeolus
-------- AMR + cobot--------
Robotnik+UR Omron+TM Clearpath+UR Kuka
National Taiwan University
Graduate Institute of Electrical Engineering
Background (2/2)
4
 Multiple modules need to be integrated and managed
 A flexible pipeline need to be constructed to complete specific scenario
Mobile robot Manipulator Sensors
Gripper Computers
Simultaneous Localization and Mapping (SLAM) Navigation
Manipulation
Grasp Pose Estimation
Object Detection
Text To Speech (TTS)
Automatic Speech Recognition (ASR)
Mobile Manipulator
National Taiwan University
Graduate Institute of Electrical Engineering
 Construct a system which can integrate and manage several functional modules
 Implement it into a consumer electronics packaging scenario with a general mobile
manipulation pipeline
Objective
5
Products
Components
X
Y
Z
Packaging Section
x
USB
Tripod
camera
Router
LiDAR
Camera
1.Grasping
2.Placing
3.Fetching
4.Stacking
Box
…….
National Taiwan University
Graduate Institute of Electrical Engineering
 We designed and developed an Autonomous Mobile Industrial Robot (AMIR)
composed of differential-drive AMR, 6 DOF industrial robot and 2 finger gripper
 With d435 RGB-D camera, 9DoF Razor IMU and two URG-04LX laser scanners
 The whole system is based on the Robot operating system (ROS) environment.
Our Robotic Platform
6
National Taiwan University
Graduate Institute of Electrical Engineering
System Architecture
7
 Our task-level architecture uses command central state machine for managing states
in a general mobile manipulation scenario based on SMACH
 The user command can be parsed to collecting and delivering
 Each function module is implemented into ROS node
National Taiwan University
Graduate Institute of Electrical Engineering 8
National Taiwan University
Graduate Institute of Electrical Engineering
SLAM & Navigation
9
 Once it gets command, the sub-state will switch between navigation and one of action
sub-states, and finally return to command state
 We use cartographer SLAM [W. Hess et al., 2016] for mapping and localization
 We use move_base for navigation
 Timed Elastic Band[C. Rösmann et al., 2017] as a local planner
 Dijkstra as a global planner
[Courtesy: http://wiki.ros.org/move_base]
National Taiwan University
Graduate Institute of Electrical Engineering
 We use Moveit! for motion planning of arm
 The general framework of the manipulation tasks is
composed of six sub-states
 PoseAttacking: move the camera above the
workspace to take a picture
 ObjectDetection: find the required object’s
bounding box and class
 PoseEstimation: estimate the grasping/stacking
pose of required object
 PickAndPlace: pick up the target object and place
to the destination
Manipulation
10
National Taiwan University
Graduate Institute of Electrical Engineering
Object Detection
11
 4 Classes: USB, Tripod, Box, Camera
 We use MobileNet-SSD [A. G. Howard et at., 2017][W. Liu et al., 2016]
for object detection implemented in TensorFlow
 It has less parameters and takes less time to
inference, while the performance can be maintained
[Courtesy: W. Liu et al., 2016]
Depthwise and pointwise
separable conv
National Taiwan University
Graduate Institute of Electrical Engineering
Grasp Pose Estimation
12
 We perform haf_grasping [D. Fischinger et al., 2015] for
point cloud based grasp pose estimation
 The filter is applied to segment the point cloud
into single object
cropped point cloud object point cloud
object grasp pose
haf_grasping
National Taiwan University
Graduate Institute of Electrical Engineering
 Background
 Objective
 System Architecture
 SLAM & Navigation
 Manipulation
 Object Detection
 Grasp Pose Estimation
 Experiment
 Demonstration
 Conclusion
Outline
13
National Taiwan University
Graduate Institute of Electrical Engineering
Experiment1 – Navigation (1/6)
14
 Our AMIR is 0.8m width and 1.2m length
 Autonomous navigating from A point in 304 room to B point in 302 room
 Avoiding obstacles and going through narrow door under prebuilt floor
map in 4 different cases.
2D occupancy grid map
National Taiwan University
Graduate Institute of Electrical Engineering
Experiment1 – Navigation (2/6)
15
 Case1: Empty corridor
National Taiwan University
Graduate Institute of Electrical Engineering
Experiment1 – Navigation (3/6)
16
 Case2: Corridor with a few obstacles
National Taiwan University
Graduate Institute of Electrical Engineering
Experiment1 – Navigation (4/6)
17
 Case3: Corridor with a few obstacles in another arrangement
National Taiwan University
Graduate Institute of Electrical Engineering
Experiment1 – Navigation (5/6)
18
 Case4: Corridor with a lot of obstacles
National Taiwan University
Graduate Institute of Electrical Engineering
Experiment1 – Navigation (6/6)
19
National Taiwan University
Graduate Institute of Electrical Engineering
Experiment2 - Object Detection (1/2)
20
 Training:
 With pretrained weights on MS COCO dataset
 Train on our dataset 320 images (4 class x 80 images)
 Testing:
 Test on our dataset 80 images (4 class x 20 images)
 If intersection over union (IOU) > 0.9: positive, else: negative
 Average Precision (AP) can be interpreted as an approximated area under
curve (AUC) of the Precision x Recall curve
𝐼𝑂𝑈 =
𝑎𝑟𝑒𝑎 𝑜𝑓 𝑜𝑣𝑒𝑟𝑙𝑎𝑝
𝑎𝑟𝑒𝑎 𝑜𝑓 𝑢𝑛𝑖𝑜𝑛
𝑃𝑟𝑒𝑐𝑖𝑠𝑖𝑜𝑛 =
𝑇𝑃
𝑇𝑃+𝐹𝑃
=
𝑇𝑃
𝑎𝑙𝑙 𝑑𝑒𝑡𝑒𝑐𝑡𝑖𝑜𝑛𝑠
𝑅𝑒𝑐𝑎𝑙𝑙 =
𝑇𝑃
𝑇𝑃+𝐹𝑁
=
𝑇𝑃
𝑎𝑙𝑙 𝑔𝑟𝑜𝑢𝑛𝑑 𝑡𝑟𝑢𝑡ℎ
National Taiwan University
Graduate Institute of Electrical Engineering
 Computing precision & recall
as the confidence is changed
 𝐴𝑃𝑈𝑆𝐵 = 100.00%
 𝐴𝑃𝐵𝑜𝑥 = 97.61%
 𝐴𝑃𝑇𝑟𝑖𝑝𝑜𝑑 = 80.00%
 𝐴𝑃𝑐𝑎𝑚𝑒𝑟𝑎 = 92.33%
 𝑚𝐴𝑃 = 92.485%
Experiment2 - Object Detection (2/2)
21
There are FPs in low confidence
Confusing between tripod and camera
There are some FNs
National Taiwan University
Graduate Institute of Electrical Engineering
 The testing flow includes three stages, camera positioning, object detection and
grasp pose estimation, and object picking and placing
 The testing flow is executed 30 times.
 In the success case, the target can be lifted from the table and transported to the destination. As the grasp is
not robust enough, that is, the grasp pose is estimated at an unstable position, this will be regarded as a semi-
success one. If the grasp fails or the target object drops during the process, the trial is labeled as failure
Experiment3 - Grasping
22
National Taiwan University
Graduate Institute of Electrical Engineering
Demonstration
23
National Taiwan University
Graduate Institute of Electrical Engineering
 We propose a finite state machine based system to manage robot behaviors using
functional modules in an industrial collecting and delivering scenario.
 Our system integrates SLAM, navigation, manipulation, object detection and
grasping pose estimation together to complete assigned tasks, making the
automation scenario more maintainable and extendable.
 Also, we successfully demonstrate our system on the AMIR based on ROS in the
real world. Each part of system is experimentally examined and obtaining great
results.
Conclusion
24
National Taiwan University
Graduate Institute of Electrical Engineering 25
Thank you :)
Q &A

More Related Content

Recently uploaded

Microscopic Analysis of Ceramic Materials.pptx
Microscopic Analysis of Ceramic Materials.pptxMicroscopic Analysis of Ceramic Materials.pptx
Microscopic Analysis of Ceramic Materials.pptxpurnimasatapathy1234
 
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...ranjana rawat
 
(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Service
(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Service(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Service
(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Serviceranjana rawat
 
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service NashikCall Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service NashikCall Girls in Nagpur High Profile
 
Introduction to Multiple Access Protocol.pptx
Introduction to Multiple Access Protocol.pptxIntroduction to Multiple Access Protocol.pptx
Introduction to Multiple Access Protocol.pptxupamatechverse
 
AKTU Computer Networks notes --- Unit 3.pdf
AKTU Computer Networks notes ---  Unit 3.pdfAKTU Computer Networks notes ---  Unit 3.pdf
AKTU Computer Networks notes --- Unit 3.pdfankushspencer015
 
HARDNESS, FRACTURE TOUGHNESS AND STRENGTH OF CERAMICS
HARDNESS, FRACTURE TOUGHNESS AND STRENGTH OF CERAMICSHARDNESS, FRACTURE TOUGHNESS AND STRENGTH OF CERAMICS
HARDNESS, FRACTURE TOUGHNESS AND STRENGTH OF CERAMICSRajkumarAkumalla
 
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete RecordCCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete RecordAsst.prof M.Gokilavani
 
Call for Papers - Educational Administration: Theory and Practice, E-ISSN: 21...
Call for Papers - Educational Administration: Theory and Practice, E-ISSN: 21...Call for Papers - Educational Administration: Theory and Practice, E-ISSN: 21...
Call for Papers - Educational Administration: Theory and Practice, E-ISSN: 21...Christo Ananth
 
Java Programming :Event Handling(Types of Events)
Java Programming :Event Handling(Types of Events)Java Programming :Event Handling(Types of Events)
Java Programming :Event Handling(Types of Events)simmis5
 
247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).ppt
247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).ppt247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).ppt
247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).pptssuser5c9d4b1
 
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...Christo Ananth
 
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur EscortsCall Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur EscortsCall Girls in Nagpur High Profile
 
UNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its PerformanceUNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its Performancesivaprakash250
 
Software Development Life Cycle By Team Orange (Dept. of Pharmacy)
Software Development Life Cycle By  Team Orange (Dept. of Pharmacy)Software Development Life Cycle By  Team Orange (Dept. of Pharmacy)
Software Development Life Cycle By Team Orange (Dept. of Pharmacy)Suman Mia
 
KubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghlyKubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghlysanyuktamishra911
 
result management system report for college project
result management system report for college projectresult management system report for college project
result management system report for college projectTonystark477637
 
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...roncy bisnoi
 
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINEMANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINESIVASHANKAR N
 
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICSAPPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICSKurinjimalarL3
 

Recently uploaded (20)

Microscopic Analysis of Ceramic Materials.pptx
Microscopic Analysis of Ceramic Materials.pptxMicroscopic Analysis of Ceramic Materials.pptx
Microscopic Analysis of Ceramic Materials.pptx
 
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
The Most Attractive Pune Call Girls Manchar 8250192130 Will You Miss This Cha...
 
(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Service
(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Service(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Service
(RIA) Call Girls Bhosari ( 7001035870 ) HI-Fi Pune Escorts Service
 
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service NashikCall Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
Call Girls Service Nashik Vaishnavi 7001305949 Independent Escort Service Nashik
 
Introduction to Multiple Access Protocol.pptx
Introduction to Multiple Access Protocol.pptxIntroduction to Multiple Access Protocol.pptx
Introduction to Multiple Access Protocol.pptx
 
AKTU Computer Networks notes --- Unit 3.pdf
AKTU Computer Networks notes ---  Unit 3.pdfAKTU Computer Networks notes ---  Unit 3.pdf
AKTU Computer Networks notes --- Unit 3.pdf
 
HARDNESS, FRACTURE TOUGHNESS AND STRENGTH OF CERAMICS
HARDNESS, FRACTURE TOUGHNESS AND STRENGTH OF CERAMICSHARDNESS, FRACTURE TOUGHNESS AND STRENGTH OF CERAMICS
HARDNESS, FRACTURE TOUGHNESS AND STRENGTH OF CERAMICS
 
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete RecordCCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
CCS335 _ Neural Networks and Deep Learning Laboratory_Lab Complete Record
 
Call for Papers - Educational Administration: Theory and Practice, E-ISSN: 21...
Call for Papers - Educational Administration: Theory and Practice, E-ISSN: 21...Call for Papers - Educational Administration: Theory and Practice, E-ISSN: 21...
Call for Papers - Educational Administration: Theory and Practice, E-ISSN: 21...
 
Java Programming :Event Handling(Types of Events)
Java Programming :Event Handling(Types of Events)Java Programming :Event Handling(Types of Events)
Java Programming :Event Handling(Types of Events)
 
247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).ppt
247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).ppt247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).ppt
247267395-1-Symmetric-and-distributed-shared-memory-architectures-ppt (1).ppt
 
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
Call for Papers - African Journal of Biological Sciences, E-ISSN: 2663-2187, ...
 
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur EscortsCall Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
Call Girls Service Nagpur Tanvi Call 7001035870 Meet With Nagpur Escorts
 
UNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its PerformanceUNIT - IV - Air Compressors and its Performance
UNIT - IV - Air Compressors and its Performance
 
Software Development Life Cycle By Team Orange (Dept. of Pharmacy)
Software Development Life Cycle By  Team Orange (Dept. of Pharmacy)Software Development Life Cycle By  Team Orange (Dept. of Pharmacy)
Software Development Life Cycle By Team Orange (Dept. of Pharmacy)
 
KubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghlyKubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghly
 
result management system report for college project
result management system report for college projectresult management system report for college project
result management system report for college project
 
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
Call Girls Pimpri Chinchwad Call Me 7737669865 Budget Friendly No Advance Boo...
 
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINEMANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
MANUFACTURING PROCESS-II UNIT-2 LATHE MACHINE
 
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICSAPPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
APPLICATIONS-AC/DC DRIVES-OPERATING CHARACTERISTICS
 

Featured

How to Prepare For a Successful Job Search for 2024
How to Prepare For a Successful Job Search for 2024How to Prepare For a Successful Job Search for 2024
How to Prepare For a Successful Job Search for 2024Albert Qian
 
Social Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie InsightsSocial Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie InsightsKurio // The Social Media Age(ncy)
 
Trends In Paid Search: Navigating The Digital Landscape In 2024
Trends In Paid Search: Navigating The Digital Landscape In 2024Trends In Paid Search: Navigating The Digital Landscape In 2024
Trends In Paid Search: Navigating The Digital Landscape In 2024Search Engine Journal
 
5 Public speaking tips from TED - Visualized summary
5 Public speaking tips from TED - Visualized summary5 Public speaking tips from TED - Visualized summary
5 Public speaking tips from TED - Visualized summarySpeakerHub
 
ChatGPT and the Future of Work - Clark Boyd
ChatGPT and the Future of Work - Clark Boyd ChatGPT and the Future of Work - Clark Boyd
ChatGPT and the Future of Work - Clark Boyd Clark Boyd
 
Getting into the tech field. what next
Getting into the tech field. what next Getting into the tech field. what next
Getting into the tech field. what next Tessa Mero
 
Google's Just Not That Into You: Understanding Core Updates & Search Intent
Google's Just Not That Into You: Understanding Core Updates & Search IntentGoogle's Just Not That Into You: Understanding Core Updates & Search Intent
Google's Just Not That Into You: Understanding Core Updates & Search IntentLily Ray
 
Time Management & Productivity - Best Practices
Time Management & Productivity -  Best PracticesTime Management & Productivity -  Best Practices
Time Management & Productivity - Best PracticesVit Horky
 
The six step guide to practical project management
The six step guide to practical project managementThe six step guide to practical project management
The six step guide to practical project managementMindGenius
 
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...RachelPearson36
 
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...Applitools
 
12 Ways to Increase Your Influence at Work
12 Ways to Increase Your Influence at Work12 Ways to Increase Your Influence at Work
12 Ways to Increase Your Influence at WorkGetSmarter
 
Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...
Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...
Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...DevGAMM Conference
 
Barbie - Brand Strategy Presentation
Barbie - Brand Strategy PresentationBarbie - Brand Strategy Presentation
Barbie - Brand Strategy PresentationErica Santiago
 
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them wellGood Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them wellSaba Software
 
Introduction to C Programming Language
Introduction to C Programming LanguageIntroduction to C Programming Language
Introduction to C Programming LanguageSimplilearn
 

Featured (20)

How to Prepare For a Successful Job Search for 2024
How to Prepare For a Successful Job Search for 2024How to Prepare For a Successful Job Search for 2024
How to Prepare For a Successful Job Search for 2024
 
Social Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie InsightsSocial Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie Insights
 
Trends In Paid Search: Navigating The Digital Landscape In 2024
Trends In Paid Search: Navigating The Digital Landscape In 2024Trends In Paid Search: Navigating The Digital Landscape In 2024
Trends In Paid Search: Navigating The Digital Landscape In 2024
 
5 Public speaking tips from TED - Visualized summary
5 Public speaking tips from TED - Visualized summary5 Public speaking tips from TED - Visualized summary
5 Public speaking tips from TED - Visualized summary
 
ChatGPT and the Future of Work - Clark Boyd
ChatGPT and the Future of Work - Clark Boyd ChatGPT and the Future of Work - Clark Boyd
ChatGPT and the Future of Work - Clark Boyd
 
Getting into the tech field. what next
Getting into the tech field. what next Getting into the tech field. what next
Getting into the tech field. what next
 
Google's Just Not That Into You: Understanding Core Updates & Search Intent
Google's Just Not That Into You: Understanding Core Updates & Search IntentGoogle's Just Not That Into You: Understanding Core Updates & Search Intent
Google's Just Not That Into You: Understanding Core Updates & Search Intent
 
How to have difficult conversations
How to have difficult conversations How to have difficult conversations
How to have difficult conversations
 
Introduction to Data Science
Introduction to Data ScienceIntroduction to Data Science
Introduction to Data Science
 
Time Management & Productivity - Best Practices
Time Management & Productivity -  Best PracticesTime Management & Productivity -  Best Practices
Time Management & Productivity - Best Practices
 
The six step guide to practical project management
The six step guide to practical project managementThe six step guide to practical project management
The six step guide to practical project management
 
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
 
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
 
12 Ways to Increase Your Influence at Work
12 Ways to Increase Your Influence at Work12 Ways to Increase Your Influence at Work
12 Ways to Increase Your Influence at Work
 
ChatGPT webinar slides
ChatGPT webinar slidesChatGPT webinar slides
ChatGPT webinar slides
 
More than Just Lines on a Map: Best Practices for U.S Bike Routes
More than Just Lines on a Map: Best Practices for U.S Bike RoutesMore than Just Lines on a Map: Best Practices for U.S Bike Routes
More than Just Lines on a Map: Best Practices for U.S Bike Routes
 
Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...
Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...
Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...
 
Barbie - Brand Strategy Presentation
Barbie - Brand Strategy PresentationBarbie - Brand Strategy Presentation
Barbie - Brand Strategy Presentation
 
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them wellGood Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
 
Introduction to C Programming Language
Introduction to C Programming LanguageIntroduction to C Programming Language
Introduction to C Programming Language
 

2020aim

  • 1. National Taiwan University Graduate Institute of Electrical Engineering Modular ROS Based Autonomous Mobile Industrial Robot System for Automated Intelligent Manufacturing Applications Ren C. Luo, Shang Lun Lee, Yu Cheng Wen, and Chin Hao Hsu Department of Electrical Engineering, National Taiwan University, Taipei, Taiwan Presenter: Shang Lun Lee 2020 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM). Boston, USA. July 9, 2020 1
  • 2. National Taiwan University Graduate Institute of Electrical Engineering  Background  Objective  Our Robotic Platform  System Architecture  SLAM & Navigation  Manipulation  Object Detection  Grasp Pose Estimation  Experiment  Demonstration  Conclusion Outline 2
  • 3. National Taiwan University Graduate Institute of Electrical Engineering Background (1/2) 3  Mobile manipulator is a growing robotic machine such as service robot and AMR+cobot  Small platform with 2~5kg arm payload v.s. large platform with 5~kg arm payload  With mobility + dexterity + perception, they can do various tasks -------- Service robot -------- Pr2 Pepper RB-1 Aeolus -------- AMR + cobot-------- Robotnik+UR Omron+TM Clearpath+UR Kuka
  • 4. National Taiwan University Graduate Institute of Electrical Engineering Background (2/2) 4  Multiple modules need to be integrated and managed  A flexible pipeline need to be constructed to complete specific scenario Mobile robot Manipulator Sensors Gripper Computers Simultaneous Localization and Mapping (SLAM) Navigation Manipulation Grasp Pose Estimation Object Detection Text To Speech (TTS) Automatic Speech Recognition (ASR) Mobile Manipulator
  • 5. National Taiwan University Graduate Institute of Electrical Engineering  Construct a system which can integrate and manage several functional modules  Implement it into a consumer electronics packaging scenario with a general mobile manipulation pipeline Objective 5 Products Components X Y Z Packaging Section x USB Tripod camera Router LiDAR Camera 1.Grasping 2.Placing 3.Fetching 4.Stacking Box …….
  • 6. National Taiwan University Graduate Institute of Electrical Engineering  We designed and developed an Autonomous Mobile Industrial Robot (AMIR) composed of differential-drive AMR, 6 DOF industrial robot and 2 finger gripper  With d435 RGB-D camera, 9DoF Razor IMU and two URG-04LX laser scanners  The whole system is based on the Robot operating system (ROS) environment. Our Robotic Platform 6
  • 7. National Taiwan University Graduate Institute of Electrical Engineering System Architecture 7  Our task-level architecture uses command central state machine for managing states in a general mobile manipulation scenario based on SMACH  The user command can be parsed to collecting and delivering  Each function module is implemented into ROS node
  • 8. National Taiwan University Graduate Institute of Electrical Engineering 8
  • 9. National Taiwan University Graduate Institute of Electrical Engineering SLAM & Navigation 9  Once it gets command, the sub-state will switch between navigation and one of action sub-states, and finally return to command state  We use cartographer SLAM [W. Hess et al., 2016] for mapping and localization  We use move_base for navigation  Timed Elastic Band[C. Rösmann et al., 2017] as a local planner  Dijkstra as a global planner [Courtesy: http://wiki.ros.org/move_base]
  • 10. National Taiwan University Graduate Institute of Electrical Engineering  We use Moveit! for motion planning of arm  The general framework of the manipulation tasks is composed of six sub-states  PoseAttacking: move the camera above the workspace to take a picture  ObjectDetection: find the required object’s bounding box and class  PoseEstimation: estimate the grasping/stacking pose of required object  PickAndPlace: pick up the target object and place to the destination Manipulation 10
  • 11. National Taiwan University Graduate Institute of Electrical Engineering Object Detection 11  4 Classes: USB, Tripod, Box, Camera  We use MobileNet-SSD [A. G. Howard et at., 2017][W. Liu et al., 2016] for object detection implemented in TensorFlow  It has less parameters and takes less time to inference, while the performance can be maintained [Courtesy: W. Liu et al., 2016] Depthwise and pointwise separable conv
  • 12. National Taiwan University Graduate Institute of Electrical Engineering Grasp Pose Estimation 12  We perform haf_grasping [D. Fischinger et al., 2015] for point cloud based grasp pose estimation  The filter is applied to segment the point cloud into single object cropped point cloud object point cloud object grasp pose haf_grasping
  • 13. National Taiwan University Graduate Institute of Electrical Engineering  Background  Objective  System Architecture  SLAM & Navigation  Manipulation  Object Detection  Grasp Pose Estimation  Experiment  Demonstration  Conclusion Outline 13
  • 14. National Taiwan University Graduate Institute of Electrical Engineering Experiment1 – Navigation (1/6) 14  Our AMIR is 0.8m width and 1.2m length  Autonomous navigating from A point in 304 room to B point in 302 room  Avoiding obstacles and going through narrow door under prebuilt floor map in 4 different cases. 2D occupancy grid map
  • 15. National Taiwan University Graduate Institute of Electrical Engineering Experiment1 – Navigation (2/6) 15  Case1: Empty corridor
  • 16. National Taiwan University Graduate Institute of Electrical Engineering Experiment1 – Navigation (3/6) 16  Case2: Corridor with a few obstacles
  • 17. National Taiwan University Graduate Institute of Electrical Engineering Experiment1 – Navigation (4/6) 17  Case3: Corridor with a few obstacles in another arrangement
  • 18. National Taiwan University Graduate Institute of Electrical Engineering Experiment1 – Navigation (5/6) 18  Case4: Corridor with a lot of obstacles
  • 19. National Taiwan University Graduate Institute of Electrical Engineering Experiment1 – Navigation (6/6) 19
  • 20. National Taiwan University Graduate Institute of Electrical Engineering Experiment2 - Object Detection (1/2) 20  Training:  With pretrained weights on MS COCO dataset  Train on our dataset 320 images (4 class x 80 images)  Testing:  Test on our dataset 80 images (4 class x 20 images)  If intersection over union (IOU) > 0.9: positive, else: negative  Average Precision (AP) can be interpreted as an approximated area under curve (AUC) of the Precision x Recall curve 𝐼𝑂𝑈 = 𝑎𝑟𝑒𝑎 𝑜𝑓 𝑜𝑣𝑒𝑟𝑙𝑎𝑝 𝑎𝑟𝑒𝑎 𝑜𝑓 𝑢𝑛𝑖𝑜𝑛 𝑃𝑟𝑒𝑐𝑖𝑠𝑖𝑜𝑛 = 𝑇𝑃 𝑇𝑃+𝐹𝑃 = 𝑇𝑃 𝑎𝑙𝑙 𝑑𝑒𝑡𝑒𝑐𝑡𝑖𝑜𝑛𝑠 𝑅𝑒𝑐𝑎𝑙𝑙 = 𝑇𝑃 𝑇𝑃+𝐹𝑁 = 𝑇𝑃 𝑎𝑙𝑙 𝑔𝑟𝑜𝑢𝑛𝑑 𝑡𝑟𝑢𝑡ℎ
  • 21. National Taiwan University Graduate Institute of Electrical Engineering  Computing precision & recall as the confidence is changed  𝐴𝑃𝑈𝑆𝐵 = 100.00%  𝐴𝑃𝐵𝑜𝑥 = 97.61%  𝐴𝑃𝑇𝑟𝑖𝑝𝑜𝑑 = 80.00%  𝐴𝑃𝑐𝑎𝑚𝑒𝑟𝑎 = 92.33%  𝑚𝐴𝑃 = 92.485% Experiment2 - Object Detection (2/2) 21 There are FPs in low confidence Confusing between tripod and camera There are some FNs
  • 22. National Taiwan University Graduate Institute of Electrical Engineering  The testing flow includes three stages, camera positioning, object detection and grasp pose estimation, and object picking and placing  The testing flow is executed 30 times.  In the success case, the target can be lifted from the table and transported to the destination. As the grasp is not robust enough, that is, the grasp pose is estimated at an unstable position, this will be regarded as a semi- success one. If the grasp fails or the target object drops during the process, the trial is labeled as failure Experiment3 - Grasping 22
  • 23. National Taiwan University Graduate Institute of Electrical Engineering Demonstration 23
  • 24. National Taiwan University Graduate Institute of Electrical Engineering  We propose a finite state machine based system to manage robot behaviors using functional modules in an industrial collecting and delivering scenario.  Our system integrates SLAM, navigation, manipulation, object detection and grasping pose estimation together to complete assigned tasks, making the automation scenario more maintainable and extendable.  Also, we successfully demonstrate our system on the AMIR based on ROS in the real world. Each part of system is experimentally examined and obtaining great results. Conclusion 24
  • 25. National Taiwan University Graduate Institute of Electrical Engineering 25 Thank you :) Q &A

Editor's Notes

  1. cited