automate - advanced robot manufacturing driving growth … manufacturing process (in review) robot...
Post on 16-Oct-2020
2 Views
Preview:
TRANSCRIPT
Advanced Robot Manufacturing Driving Growth through Increased
Production Flexibility
Larry M. SweetGeorgia Tech Institute for Robotics and Intelligent Machines
Explosive Robot Growth through 2025
• 11% CAGR unit growth through 2025
Source: BCG report “The Robot Revolution”, September 2015
• Cost reductions: • Robot HW / SW, perception• System engineering• Safety systems
0
10
20
30
40
50
60
70
80
2015 2020 2025
Global Robot Market, $B
Military
Industrial
Commercial
Personal
0
20
40
60
80
100
120
140
2015 2020 2025
Typical Total System Costs, $K
Robot
Peripherals
System engineering
Project management
Potential for Automation
Source: McKinsey & Company, “The Technical Potential for Automation in the U.S.”, 2016
D
D
Growth is Top Priority for Industry
• OEM’s, Tier 1 suppliers, SME’s all see growth as priority
• Increased flexibility: increase capacity for new customers and products
• Major positive economic impact through the entire supply chain
Will Robots Provide Sufficient Flexibility?
How to produce higher product mix demanded by customers?
How to delivery when they want it (and return it too)?
Advanced Robot Manufacturing Priorities
• Increase production rates 30% to 50%
• Mass manufacturing -> mass customization
• Multiple manufacturing processes
• Reduce cost of integration
• Humans and co-robots working in limited floor space
Potential for 30% to 50% Gains
-
0.5
1.0
1.5
2.0
Prior New
-
0.5
1.0
1.5
2.0
Prior New
-
0.5
1.0
1.5
2.0
Current New
Consumer goods(in production)
Automated DC(in production)
Aerospace(opportunity)
High Impact Emerging Technologies
Topics for today:• Autonomous mobility
• Fixtureless manufacturing
• Collaborative robots
• Virtual product / process development
Flexible Consumer Goods Production
• New material flow, food processes, controls, robotics
• All products, flavors, sizes on each line
• 50% improvement in asset utilization
• Sales growth through increased product mix
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
Production hours
Lin
esLi
nes
Optimized flexible system
Legacy system
50% boost
Why Are Mobile Robots Disruptive?
• New levels of performance• Transactions per hour
• Sequencing
• Storage density
• Easy to re-configure• Change in SKU mix or velocity
• Change in business model
• New algorithm challenges• Replace fixed hardware with flexible
software
• Leverage research on autonomous mobile robots
Sequencing with Autonomous Robots
1
3
5
2
4
12345Output sequence 1
3
5
2
4
1 12345Output sequence 2
5
3 1
2
4
54321Input sequence 1
1 2 3 4 5 6 7 8
Production order: 395 SKU, 476 cases
Aisle number
Fully Automated Distribution Centers
• DC’s for retail, grocery, food service, beverage
• 40% gain in storage density, sequenced output for mixed palletizing
• Reduce need for capital expansion, largely pays for automation
• Locate closer to urban centers for same day delivery
• Five to ten leading integrators moving to mobile robot technology
Aerospace Fixed Assets
Increase asset utilization• More units / month
• Lower fixed costs / unit
• Lower changeover costs
Painting
InspectionAssembly
Drilling
Composites
Flexibility Options
Collaborative robots Mobile robotsIndustrial robots
Flexible fixtures No fixtures
Assembly with Shims and Positioners
Aircraft assembly shims
Composite structure
Adjustable assembly positioners
Sources: Aviation Week, April 17, 2015; Duqueine Group, Airbus A350; Assembly Magazine, October 3, 2014..
3D Surface Scanning of Workpiece
• Measurements of mating surfaces
• Laser tracking
• Laser scanning
• Photogrammetry
• Machine excess material to net shape
Source: “ Interface Management in Wing Box Assembly”, B. Chuovion, A. Popov, S. Ratchev, C. Mason, M.
Summers, SAE 2011-01-2640.
Cutting path to eliminate clash between rib feet and cover (U. Nottingham, Airbus)
3D workpiece surface scan (Georgia Tech)
Fixtureless Aerospace Manufacturing
Technical challenges• Robot absolute position accuracy
• Robot compliance subject to machining forces
• Resonance excitation during machining
• Variable depth of cut due to workpiece tolerances
Future vision
Flexible Robotic Machining Technology Roadmap
• Robot, workpiece, fixture compliance models
• Redundancy resolution to maximize stiffness
• Real-time robot control
Shreyes add sensor photo here
Robot complianceMetrology Process knowledge
• Workpiece and feature
location in world, robot
coordinates
• 3D surface scanning
• Secondary encoders
• Wireless sensing for in-situ process monitoring
• Process modeling
• Sensor and process model integration into robot control
Metrology and Enhanced Accuracy
ElectroimpactTM contributions:• Role of laser tracking of EOAT for absolute accuracy
• Detailed kinematic and stiffness models
• Secondary encoders
• Demonstration of accurate drilling under +/- 0.125 mm
Source:. “Applied Accurate Robotic Drilling for Aircraft Fuselage”, R. DeVlieg and T. Szallay,
SAE Paper 2010-01-1836.
Laser tracking EOAT Joint secondary encoder Laser tracking EOAT
Compliance Optimization
• Detailed measurements of joint and link static stiffness
• Redundancy resolution exploits extra degrees-of-freedom to find optimal stiffness
• Optimization should be task specific, along vector of machining forces
Source: “Joint stiffness identification of six-revolute industrial serial robots”, C. Dumas, S. Caro, S. Garnier, B.
Furet, Robotics and Computer-Integrated Manufacturing 27 (2011) 881–888.
Source: T. Cvitanic, V. Nguyen, L. Sweet, Collaborative and Advanced Robotic Manufacturing Lab, Georgia Tech
Stiffness Optimization for Milling
Fraunhofer Institute (IPA), TU Stuttgart contributions:• Non-linear joint compliance experiments
• Redundancy resolution for optimal stiffness
• Open loop machining yields significant tolerance variation
Source: Julian Ricardo Diaz Posada, Ulrich Schneider, Arjun Sridhar and Alexander Verl, “Automatic Motion
Generation for Robotic Milling Optimizing Stiffness with Sample-Based Planning”, MPDI Machines, 2017, 5, 3.
Machining Process Sensor Feedback
Low cost, thin film PVDF wireless sensing for in-situ process monitoring of cutting forces up to 10 MHz
Machining process modeling
Sensor and process model integration into robot control yields > 70% error reduction
Wireless tool force sensor Undercut coordinates CMM measurements
Source: “A Wireless Force-Sensing and Model-Based Approach for Enhancement of Machining Accuracy in Robotic
Milling”, L.Cen, S.Melkote, J. Castle, H. Appelman, IEEE/ASME Trans. on Mechatronics, vol. 21, no. 5, Oct 2016.
Dynamics and Chatter Control
Georgia Tech contributions:• Congruent Convergent Transformation (CCT) based mode coupling chatter
avoidance
• Expands range of permissible tool feed orientation or workpiece orientation
Source: “ CCT-based Mode Coupling Chatter Avoidance in Robotic Milling ”, L.Cen, S.Melkote, Journal of
Manufacturing Process (in review)
Robot mounted milling tool Chatter avoidance demonstration
Roadmap to Accurate Machining
Future of Collaborative Manufacturing
Aerospace industry example
Standards for Safe Human Interaction
• ISO/TS 15066 Technical Specification• Safety-rated monitored stop
• Hand guiding
• Speed and separation monitoring
• Power and force limiting
• Still requires full risk assessment • Includes all equipment and processes in
cell
• following industry best practices (FMECA)
Series elastic actuators (ReThink)
Current / encoder based torque estimates (Universal)
Pressure sensitive skin (Kuka)
Load cell in base (Fanuc)
Power and force limiting approaches
Sequential Process Example
Collaborative Operations
• Collaborative operations• Robot and human perform tasks
simultaneously during task execution
• Sequential process flow• Most common implementation today
• Human and CoBot have separate tasks
• Buffers in between to prevent starving / blocking
• Task sharing• Human and CoBot work at team
• Requires higher level of coordination, communication, flexibility
Extending CoBot Perception to 3D
Part flat on table Part tilted
2D vision pick limited to parts on flat surface 3D vision tracking on flexible conveyor
3D Vision Tracking for Moving Targets
Robot Operating System (ROS) Enabled Integration
Universal Robot Kuka KR 500 ReThink Robot Kuka KR 210 (3)
Laser tracker / scanner Tool force sensor Milling head 3D vision tracker
Liquid cooled GPU-based controller
Collaborative and Advanced Robotic Manufacturing Lab (Georgia Tech)
Hardware Robot Operating System (H-ROS)
Source: “The shift in the robotics paradigm — the Hardware Robot Operating System (H-ROS), an infrastructure to
create interoperable robot components”, V. Mayoral, A. Hernandez, R. Kojcev, I. Muguruza, I. Zamalloa, L. Usategi,
A. Bilbao.
Multi-Level Human Collaboration
• Planning, scheduling, flow
• Re-planningTask level
• Robot path / process planning
• Mobile robot route planning
• Adaptive controlControl level
• Machines, processes, fixtures
• Machine, process, perceptionMachine level
Planning and control levels Human interaction
Contact
Larry M. Sweet
Associate Director
Georgia Institute of Technology
Institute for Robotics and Intelligent Machines
Collaborative and Advanced Robotic
Manufacturing Laboratory
Atlanta, Georgia
USA
770.862.1953
larry.sweet@cc.gatech.edu
www.gtcarmlab.net
top related