Robotic arm design
Download
1 / 97

Robotic Arm Design - PowerPoint PPT Presentation


  • 581 Views
  • Updated On :

Robotic Arm Design. Joseph T. Wunderlich, Ph.D. “Lunar Roving Vehicle” (LRV). Robotic Arms. Human Arms do the dexterous “manipulation” work on manned missions.

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

PowerPoint Slideshow about 'Robotic Arm Design' - bernad


An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
Robotic arm design l.jpg

Robotic Arm Design

Joseph T. Wunderlich, Ph.D.


Slide2 l.jpg

“Lunar Roving Vehicle” (LRV)

Robotic Arms

Human Arms do the dexterous “manipulation” work on manned missions

Image from: Young, A.H. Lunar and planetary rovers: the wheels of Apollo and the quest for mars, Springer; 1 edition, August 1, 2006.


Slide3 l.jpg

“Lunar Roving Vehicle” (LRV)

Robotic Arms

1971

Human Arms do the dexterous “manipulation” work on manned missions

Image from: Young, A.H. Lunar and planetary rovers: the wheels of Apollo and the quest for mars, Springer; 1 edition, August 1, 2006.


Mars rovers l.jpg

Robotic Arms

Mars Pathfinder “Sojourner”

Mars Rovers

1996

“APXS Deployment Mechanism”

(Robotic Arm)

Image from: http://marsprogram.jpl.nasa.gov/MPF/mpf/sci_desc.html

Alpha Proton X-Ray Spectrometer


Mars rovers5 l.jpg

Robotic Arms

“Spirit” & “Opportunity”

Mars Rovers

2004

Image from: Young, A.H. Lunar and planetary rovers: the wheels of Apollo and the quest for mars, Springer; 1 edition, August 1, 2006.

“Instrument

Deployment

Device”

(Robotic Arm)


Mars rovers6 l.jpg

Robotic Arms

“Spirit” & “Opportunity”

Mars Rovers

2004

“Instrument

Deployment

Device”

(Robotic Arm)

Image from: Young, A.H. Lunar and planetary rovers: the wheels of Apollo and the quest for mars, Springer; 1 edition, August 1, 2006.


Mars rovers7 l.jpg

Robotic Arms

“Spirit” & “Opportunity”

Mars Rovers

2004

Image from: Young, A.H. Lunar and planetary rovers: the wheels of Apollo and the quest for mars, Springer; 1 edition, August 1, 2006.

“Instrument

Deployment

Device”

(Robotic Arm)


Mars rovers8 l.jpg

Robotic Arms

“Spirit” & “Opportunity”

Mars Rovers

2004

“Instrument

Deployment

Device”

(Robotic Arm)

Image from: Young, A.H. Lunar and planetary rovers: the wheels of Apollo and the quest for mars, Springer; 1 edition, August 1, 2006.


Mars rovers9 l.jpg

Robotic Arms

Mars Science Lab

Mars Rovers

2000’s

Robotic Arm

Image from: http://nssdc.gsfc.nasa.gov/planetary/mars_future.html


Mars rovers10 l.jpg

Robotic Arms

Mars Rovers

ESA “ExoMars” Rover Concept

2000’s and 2010’s

Robotic Arm

Image from: http://science.au.dk/nyheder-og-arrangementer/nyhed/article/dansk-deltagelse-i-marsforskning/


Mars rovers11 l.jpg

Robotic Arms

ESA

“ExoMars”

Rover

Concept

Mars Rovers

2000’s and 2010’s

Image from: http://www.nasa.gov/centers/jpl/news/urey-20070209.html

It has a drill


Kinematics review l.jpg

Robotic Arms

Kinematics review

Before studying advanced robotic arm design we need to review basic manipulator kinematics. So let’s look at pages 9 and 10 of:

Wunderlich, J.T. (2001).Simulation vs. real-time control; with applications to robotics and neural networks. In Proceedings of 2001 ASEE Annual Conference & Exposition, Albuquerque, NM: (session 2793), [CD-ROM]. ASEE Publications.


Kinematics review13 l.jpg

Robotic Arms

Kinematics review

FROM: Wunderlich, J.T. (2001).Simulation vs. real-time control; with applications to robotics and neural networks. In Proceedings of 2001 ASEE Annual Conference & Exposition, Albuquerque, NM: (session 2793), [CD-ROM]. ASEE Publications.


Kinematics review14 l.jpg

Robotic Arms

Kinematics review

FROM: Wunderlich, J.T. (2001).Simulation vs. real-time control; with applications to robotics and neural networks. In Proceedings of 2001 ASEE Annual Conference & Exposition, Albuquerque, NM: (session 2793), [CD-ROM]. ASEE Publications.


Slide15 l.jpg

Robotic Arms

What kind of arm has OPTIMAL DEXTERITY?A Redundant Manipulator? (i.e., MoreDegrees Of Freedom than you need)

Human Arm is a 7 DOF Redundant Manipulator

3 DOF

3 DOF

1 DOF


Slide16 l.jpg

Robotic Arms

What kind of arm has OPTIMAL DEXTERITY?A Hyper-Redundant Manipulator? (i.e., Many moreDegrees Of Freedom than “needed”)


Slide17 l.jpg

Robotic Arms

OPTIMAL DEXTERITY?Would manyHyper-Redundant Manipulators be optimal? (i.e., Each with many moreDegrees Of Freedom than “needed”)


J wunderlich related publications l.jpg
J. Wunderlich Related Publications

Wunderlich, J.T. (2004).Simulating a robotic arm in a box: redundant kinematics, path planning, and rapid-prototyping for enclosed spaces. In Transactions of the Society for Modeling and Simulation International: Vol. 80. (pp. 301-316). San Diego, CA: Sage Publications.

Wunderlich, J.T. (2004).Design of a welding arm for unibody automobile assembly. In Proceedings of IMG04 Intelligent Manipulation and Grasping International Conference, Genova, Italy, R. Molfino (Ed.): (pp. 117-122). Genova, Italy: Grafica KC s.n.c Press.








Simulation initialization l.jpg
Simulation (initialization)


Simulation go to task start point l.jpg
Simulation (go to task start point)


Simulation perform welding task l.jpg
Simulation (perform welding task)


Path planning l.jpg
Path Planning

  • Pseudo-inverse velocity control

  • Attractive poles

  • Repelling fields


Path planning39 l.jpg
Path Planning

  • Pseudo-inverse velocity control

by specifying a desired end-effector velocity and a desired obstacle-avoidance-point velocity

For derivation of these equations, read pages 1 to 4 of:

Wunderlich, J.T. (2004).Simulating a robotic arm in a box: redundant kinematics, path planning, and rapid-prototyping for enclosed spaces. In Transactions of the Society for Modeling and Simulation International: Vol. 80. (pp. 301-316). San Diego, CA: Sage Publications.


Path planning40 l.jpg
Path Planning

  • Pseudo-inverse velocity control

    • With new proposed methodology here:

by specifying a desired end-effector velocity and multiple desired obstacle-avoidance-point velocities:

For derivation of these equations, read pages 1 to 4 of:

Wunderlich, J.T. (2004).Simulating a robotic arm in a box: redundant kinematics, path planning, and rapid-prototyping for enclosed spaces. In Transactions of the Society for Modeling and Simulation International: Vol. 80. (pp. 301-316). San Diego, CA: Sage Publications………………………


Path planning41 l.jpg

FROM: Wunderlich, J.T. (2004).Simulating a robotic arm in a box: redundant kinematics, path planning, and rapid-prototyping for enclosed spaces. In Transactions of the Society for Modeling and Simulation nternational: Vol. 80. (pp. 301-316). San Diego, CA: Sage Publications

Path Planning


Path planning42 l.jpg

FROM: Wunderlich, J.T. (2004).Simulating a robotic arm in a box: redundant kinematics, path planning, and rapid-prototyping for enclosed spaces. In Transactions of the Society for Modeling and Simulation nternational: Vol. 80. (pp. 301-316). San Diego, CA: Sage Publications

Path Planning


Path planning43 l.jpg

FROM: Wunderlich, J.T. (2004).Simulating a robotic arm in a box: redundant kinematics, path planning, and rapid-prototyping for enclosed spaces. In Transactions of the Society for Modeling and Simulation nternational: Vol. 80. (pp. 301-316). San Diego, CA: Sage Publications

Path Planning


Path planning44 l.jpg
Path Planning

  • Pseudo-inverse velocity control

    • Technique made feasible by:

      • Attractive Poles

      • Repelling Fields

        • Proportional to obstacle proximity

        • Direction related to poles (or goal)

        • Limited range


Search for feasible designs l.jpg
Search for Feasible Designs

1) Guess initial kinematics

  • Link-lengths and DOF to reach furthest point in unibody

    2) Find repelling-velocity magnitudes

    3) Use heuristic(s) to change link-lengths

  • Test new designs

  • Can minimize DOF directly


Example search l.jpg
Example Search

  • Using heuristic that changes link lengths by 10cm, two at a time, results in 3489 new designs from an original (90,120,95,50,40,)cm 5-DOF design

    • This includes 104 4-DOF designs

  • Another search; one designed specifically to minimize DOF, quickly yields 15 4-DOF designs (and 41 5-DOF designs)


New 4 dof design generated from original 5 dof design l.jpg
New 4-DOF Design (Generated from original 5-DOF design)


New 4 dof design generated from original 5 dof design48 l.jpg
New 4-DOF Design (Generated from original 5-DOF design)


New 4 dof design generated from original 5 dof design49 l.jpg
New 4-DOF Design (Generated from original 5-DOF design)


New 4 dof design generated from original 5 dof design50 l.jpg
New 4-DOF Design (Generated from original 5-DOF design)


New 4 dof design generated from original 5 dof design51 l.jpg
New 4-DOF Design (Generated from original 5-DOF design)


New 4 dof design generated from original 5 dof design52 l.jpg
New 4-DOF Design (Generated from original 5-DOF design)


New 4 dof design generated from original 5 dof design53 l.jpg
New 4-DOF Design (Generated from original 5-DOF design)


New 4 dof design generated from original 5 dof design54 l.jpg
New 4-DOF Design (Generated from original 5-DOF design)


New 4 dof design generated from original 5 dof design55 l.jpg
New 4-DOF Design (Generated from original 5-DOF design)


New 4 dof design generated from original 5 dof design56 l.jpg
New 4-DOF Design (Generated from original 5-DOF design)


New 4 dof design generated from original 5 dof design57 l.jpg
New 4-DOF Design (Generated from original 5-DOF design)


New 4 dof design generated from original 5 dof design58 l.jpg
New 4-DOF Design (Generated from original 5-DOF design)


Selecting a design l.jpg
Selecting a Design

  • Compare measures taken during search

    • DOF

    • Joint-angle displacement

    • Manipulability

    • Simulated speed

    • Consumption of available redundancy


Dof minimize l.jpg
DOF (minimize)

  • Decrease initial financial cost

  • Decrease financial operating costs


Joint angle displacement minimize l.jpg
Joint-angle displacement (minimize)

  • Related to the mechanical work required to maneuver

    • Increase usable life of equipment

    • Decrease financial operating costs


Manipulability maximize l.jpg
Manipulability (maximize)

  • Indication of how far arm configuration is from singularities over trajectory


New proposed measure here c onsumption o f a vailable r edundancy coar minimize it l.jpg
New proposed measure here:Consumption Of Available Redundancy (COAR)(minimize it)

  • Indication how redundancy used over trajectory

  • COAR varies significantly over trajectory when joint angle changes vary significantly due to obstacle avoidance


Coar example highly constricted workspace l.jpg
COAR(Example highly-constrictedworkspace)


Simulated speed maximize l.jpg
Simulated speed (maximize)

  • Simply number of simulation steps in a trajectory

    • Indication of how trajectory compromised

      • Local minima

      • High COAR


High quality final design selected from all feasible designs from search l.jpg
“High-Quality” Final Design (selected from all feasible designs from search)


How robust is methodology l.jpg
How Robust is Methodology?

  • Dependent on initial configuration?

  • Can escape local minima?

  • Can deal with singularities?

  • Can be extended to more complex workspaces?









Can deal with singularities l.jpg
Can deal with singularities?

  • Considered “damped least squares” and “weighting matrix”

  • Considered treating singularity configurations as obstacles

    • but could push arm into repelling fields

  • Using “manipulability measure” to compare all candidate designs

    • “natural selection”


Create enclosure from simulation primitives l.jpg

Extend methodology to more complex workspaces

Create enclosure from simulation Primitives


How robust is methodology79 l.jpg
How Robust is Methodology?

  • Dependent on initial configuration?

    • Only somewhat

  • Can escape local minima?

    • Yes

  • Can deal with singularities?

    • Considered less desirable designs

  • Can be extended to more complex workspaces?

    • Yes




Slide82 l.jpg

Results of a test design run where red arms are successful at reaching goal (red X) and blue arms are not.

Circles show elbows being repelled from surfaces


Slide83 l.jpg

Robotic Arm Design at reaching goal (red X) and blue arms are not.

Using complex path-planning and obstacle avoidance

Results of a test design run where red arms are successful at reaching goal (red X) and blue arms are not.

Circles show elbows being repelled from surfaces


Slide84 l.jpg

Robotic Arm Design at reaching goal (red X) and blue arms are not.

Selected Design from Search

Wunderlich, J.T. (2004).Simulating a robotic arm in a box: redundant kinematics, path planning, and rapid-prototyping for enclosed spaces. In Transactions of the Society for Modeling and Simulation International: Vol. 80. (pp. 301-316). San Diego, CA: Sage Publications.


Slide85 l.jpg

Robotic Arm Design at reaching goal (red X) and blue arms are not.

  • This modified psuedoinverse path-planning works fine for rapidly prototyping designs

    • And can easily use simpler control scheme for real-time control if concerned about psuedoinverse velocity control implementation difficulties

  • Rapid prototypingof quality designs

    • Dexterous

    • Minimal DOF

    • Low energy

    • Good geometric fit

    • Semi task-specific


Future possibilities l.jpg

Robotic Arm Design at reaching goal (red X) and blue arms are not.

Future Possibilities

  • 3-D

    • Tubular primitives

    • Cube primitives

  • Dynamic Model to optimize forces for cutting, drilling, material handling, etc.

  • Learn environment (anticipate walls)

  • Adaptive repelling fields

  • Use COAR to drive design process

  • Probabilistically complete search

  • Extend methodology to design of a manipulator for an Aerospace application …………………………


2020 esa nasa europa jupiter system mission ejsm l.jpg

Robotic Arms at reaching goal (red X) and blue arms are not.

2020 ESA/NASAEuropa Jupiter System Mission”(EJSM)

“A Joint International MissionThe baseline EJSM consists of two primary flight elements operating in the Jovian system: the NASA-led Jupiter Europa Orbiter (JEO) , and the ESA-led Jupiter Ganymede Orbiter (JGO) . JEO and JGO will execute a choreographed exploration of the Jupiter System before settling into orbit around Europa and Ganymede, respectively. JEO and JGO carry 11 and 10 complementary instruments, respectively, to monitor dynamic phenomena (such as Io’s volcanoes and Jupiter’s atmosphere), map the Jovian magnetosphere and its interactions with the Galilean satellites, and characterize water oceans beneath the ice shells of Europa and Ganymede. “

SOURCE: http://opfm.jpl.nasa.gov/europajupitersystemmissionejsm/


Europa rover an optional course project concept paper l.jpg

Robotic Arms at reaching goal (red X) and blue arms are not.

Europa RoverAn optional course project Concept Paper

  • The mission objective is to explore an ocean confirmed in 2025 to be under the ice of Europa. Assume your launch is scheduled for 2040.

  • Also assume one of the following:

  • The EuropaJupiter System Mission discovers some very thin patches of ice (less than 200 meters thick) created by localized sub-surface thermal anomalies.

  • OR

  • A mission concurrent to yours (but designed by others) has created craters on Europa’s surface that have frozen over with approximately 200 meters of ice; but assume the ice will quickly freeze much thicker -- and therefore a rapid execution of all mission operations is critical.


Europa rover l.jpg

Robotic Arms at reaching goal (red X) and blue arms are not.

Europa Rover

Your rover must be able to:

- Maneuver on icy surface

- Drill through 200 meters of ice

- When liquid water reached, either:

(1) Act as a UUV, or

(2) Deploy 100 small (10 cm.) networked UUV’s

- Communicate with UUV’s if option (2) chosen

- Communicate with base station that is communicating with several

orbiters, and earth; and is also running a concurrent simulation building

an environmental map of the region of Europa being explored. Simulation

information should also be communicated back to the rover, and then to

UUV’s if option (2) chosen; this is to help with exploration, and

preservation of the rover.

- Optionally, control a hyper-redundant manipulator attached to the rover to

aid with exploration, digging, and/or deployment of small UUV’s

- Withstand extremely cold temperatures (-143C, -225F max)

- Power itself by energy source other than sun since incident solar

radiation reaching Europa is minimal; propose a means of powering the

rover.


Europa rover90 l.jpg
Europa Rover at reaching goal (red X) and blue arms are not.

Image from http://www.newscientist.com/article/dn2929-thin-ice-opens-lead-for-life-on-europa.html


Europa rover an optional course project concept paper91 l.jpg
Europa Rover at reaching goal (red X) and blue arms are not.An optional course project Concept Paper

Read

Course Syllabus

for more details

Image from: http://www.mapaplanet.org/explorer/help/data_set.html


Optimal dexterity would many hyper redundant manipulators be optimal l.jpg

Robotic Arms at reaching goal (red X) and blue arms are not.

OPTIMAL DEXTERITYWould manyHyper-Redundant Manipulators be optimal?


Slide94 l.jpg

Robotic Arms at reaching goal (red X) and blue arms are not.

CLASS DISCUSSION OF HANDOUTS(e.g. large photocopied packet distributed on 3rd day of U.Trento lectures)

From HANDOUT including excerpts from:

S. .B. Niku, Introduction to Robotics: Analysis, Systems, Applications, Prentice Hall, July 30, 2001.


Slide95 l.jpg

Robotic Arms at reaching goal (red X) and blue arms are not.

CLASS DISCUSSION OF HANDOUTS(e.g. large photocopied packet distributed on 3rd day of U.Trento lectures)

From HANDOUT of 1992 J. Wunderlich talk at A.I. Dupont Research Institute including excerpts from several advanced robotics texts


Slide96 l.jpg

Robotic Arms at reaching goal (red X) and blue arms are not.

CLASS DISCUSSION OF HANDOUTS(e.g. large photocopied packet distributed on 3rd day of U.Trento lectures)

From HANDOUT of 1992 J. Wunderlich talk at A.I. Dupont Research Institute including excerpts from several advanced robotics texts


Slide97 l.jpg

Robotic Arms at reaching goal (red X) and blue arms are not.

Read more on telerobotics and force-feedback in:Wunderlich, J.T., S. Chen, D. Pino, and T. Rahman (1993). Software architecture for a kinematically dissimilar master-slave telerobot. In Proceedings of SPIE Int'l Conference on Telemanipulator Technology and Space Telerobotics, Boston, MA: Vol. (2057). (pp. 187-198). SPIE Press.