Neural-network-based contact force observers for haptic applications
Identifieur interne : 006061 ( Main/Exploration ); précédent : 006060; suivant : 006062Neural-network-based contact force observers for haptic applications
Auteurs : Andrew C. Smith [Canada] ; Farid Mobasser [Canada] ; Keyvan Hashtrudi-Zaad [Canada]Source :
- IEEE transactions on robotics [ 1552-3098 ] ; 2006.
Descripteurs français
- Pascal (Inist)
- Identification système, Observateur, Robotique, Système homme machine, Téléopération, Système dynamique, Système incertain, Homme, Régulation boucle multiple, Sensibilité tactile, Bilatéral, Capteur force, Mesure force, Mesure couple, Inertie, Largeur bande, Instabilité, Main, Dispositif planaire, Réseau neuronal, Modèle entrée sortie, Erreur quadratique moyenne, Pantographe.
- Wicri :
- topic : Observateur, Robotique, Homme.
English descriptors
- KwdEn :
- Bandwidth, Bilateral, Dynamical system, Force measurement, Force transducer, Hand, Human, Inertia, Input output model, Instability, Man machine system, Mean square error, Multiple loop regulation, Neural network, Observer, Pantograph, Planar device, Remote operation, Robotics, System identification, Tactile sensitivity, Torque measurement, Uncertain system.
Abstract
In the majority of robotic and haptic applications, including manipulation and human-robot interaction, contact force needs to be monitored and controlled. Transparent implementation of bilateral teleoperation or haptic controllers necessitates the exchange of operator and environment contact forces. This requires the use of expensive commercially available force/torque sensors, which are rather bulky, are vulnerable to impact forces, and increase system inertia and compliance. An alternative solution is the use of dynamic force observers, which estimate external forces using system dynamic model. However, due to the uncertainties in system dynamic structure and parameters, these model-based observers do not produce accurate force estimates, and often create a dynamic lag that may cause bandwidth limitation and instability. This paper proposes two neural-network-based force/torque observers that do not require a system dynamic model. The observers can estimate human hand force and environment contact force with up to 98.3% accuracy in the sense of mean-square error, and with negligible dynamic lag. The performance of the proposed observers are extensively analyzed in separate human-robot and robot-environment experimental settings, and in a two-channel bilateral teleoperation control loop with multiple runs with two Planar Twin-Pantograph haptic devices.
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream PascalFrancis, to step Corpus: 000B68
- to stream PascalFrancis, to step Curation: 000927
- to stream PascalFrancis, to step Checkpoint: 000957
- to stream Main, to step Merge: 006314
- to stream Main, to step Curation: 006061
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Neural-network-based contact force observers for haptic applications</title>
<author><name sortKey="Smith, Andrew C" sort="Smith, Andrew C" uniqKey="Smith A" first="Andrew C." last="Smith">Andrew C. Smith</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Robotics and Computer Vision Laboratory, Department of Electrical and Computer Engineering, Queen's University</s1>
<s2>Kingston, ON K7L 3N6</s2>
<s3>CAN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Canada</country>
<wicri:noRegion>Kingston, ON K7L 3N6</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Mobasser, Farid" sort="Mobasser, Farid" uniqKey="Mobasser F" first="Farid" last="Mobasser">Farid Mobasser</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Robotics and Computer Vision Laboratory, Department of Electrical and Computer Engineering, Queen's University</s1>
<s2>Kingston, ON K7L 3N6</s2>
<s3>CAN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Canada</country>
<wicri:noRegion>Kingston, ON K7L 3N6</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Hashtrudi Zaad, Keyvan" sort="Hashtrudi Zaad, Keyvan" uniqKey="Hashtrudi Zaad K" first="Keyvan" last="Hashtrudi-Zaad">Keyvan Hashtrudi-Zaad</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Robotics and Computer Vision Laboratory, Department of Electrical and Computer Engineering, Queen's University</s1>
<s2>Kingston, ON K7L 3N6</s2>
<s3>CAN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Canada</country>
<wicri:noRegion>Kingston, ON K7L 3N6</wicri:noRegion>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">INIST</idno>
<idno type="inist">07-0201416</idno>
<date when="2006">2006</date>
<idno type="stanalyst">PASCAL 07-0201416 INIST</idno>
<idno type="RBID">Pascal:07-0201416</idno>
<idno type="wicri:Area/PascalFrancis/Corpus">000B68</idno>
<idno type="wicri:Area/PascalFrancis/Curation">000927</idno>
<idno type="wicri:Area/PascalFrancis/Checkpoint">000957</idno>
<idno type="wicri:doubleKey">1552-3098:2006:Smith A:neural:network:based</idno>
<idno type="wicri:Area/Main/Merge">006314</idno>
<idno type="wicri:Area/Main/Curation">006061</idno>
<idno type="wicri:Area/Main/Exploration">006061</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Neural-network-based contact force observers for haptic applications</title>
<author><name sortKey="Smith, Andrew C" sort="Smith, Andrew C" uniqKey="Smith A" first="Andrew C." last="Smith">Andrew C. Smith</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Robotics and Computer Vision Laboratory, Department of Electrical and Computer Engineering, Queen's University</s1>
<s2>Kingston, ON K7L 3N6</s2>
<s3>CAN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Canada</country>
<wicri:noRegion>Kingston, ON K7L 3N6</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Mobasser, Farid" sort="Mobasser, Farid" uniqKey="Mobasser F" first="Farid" last="Mobasser">Farid Mobasser</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Robotics and Computer Vision Laboratory, Department of Electrical and Computer Engineering, Queen's University</s1>
<s2>Kingston, ON K7L 3N6</s2>
<s3>CAN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Canada</country>
<wicri:noRegion>Kingston, ON K7L 3N6</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Hashtrudi Zaad, Keyvan" sort="Hashtrudi Zaad, Keyvan" uniqKey="Hashtrudi Zaad K" first="Keyvan" last="Hashtrudi-Zaad">Keyvan Hashtrudi-Zaad</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Robotics and Computer Vision Laboratory, Department of Electrical and Computer Engineering, Queen's University</s1>
<s2>Kingston, ON K7L 3N6</s2>
<s3>CAN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Canada</country>
<wicri:noRegion>Kingston, ON K7L 3N6</wicri:noRegion>
</affiliation>
</author>
</analytic>
<series><title level="j" type="main">IEEE transactions on robotics </title>
<title level="j" type="abbreviated">IEEE trans. robot. </title>
<idno type="ISSN">1552-3098</idno>
<imprint><date when="2006">2006</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><title level="j" type="main">IEEE transactions on robotics </title>
<title level="j" type="abbreviated">IEEE trans. robot. </title>
<idno type="ISSN">1552-3098</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Bandwidth</term>
<term>Bilateral</term>
<term>Dynamical system</term>
<term>Force measurement</term>
<term>Force transducer</term>
<term>Hand</term>
<term>Human</term>
<term>Inertia</term>
<term>Input output model</term>
<term>Instability</term>
<term>Man machine system</term>
<term>Mean square error</term>
<term>Multiple loop regulation</term>
<term>Neural network</term>
<term>Observer</term>
<term>Pantograph</term>
<term>Planar device</term>
<term>Remote operation</term>
<term>Robotics</term>
<term>System identification</term>
<term>Tactile sensitivity</term>
<term>Torque measurement</term>
<term>Uncertain system</term>
</keywords>
<keywords scheme="Pascal" xml:lang="fr"><term>Identification système</term>
<term>Observateur</term>
<term>Robotique</term>
<term>Système homme machine</term>
<term>Téléopération</term>
<term>Système dynamique</term>
<term>Système incertain</term>
<term>Homme</term>
<term>Régulation boucle multiple</term>
<term>Sensibilité tactile</term>
<term>Bilatéral</term>
<term>Capteur force</term>
<term>Mesure force</term>
<term>Mesure couple</term>
<term>Inertie</term>
<term>Largeur bande</term>
<term>Instabilité</term>
<term>Main</term>
<term>Dispositif planaire</term>
<term>Réseau neuronal</term>
<term>Modèle entrée sortie</term>
<term>Erreur quadratique moyenne</term>
<term>Pantographe</term>
</keywords>
<keywords scheme="Wicri" type="topic" xml:lang="fr"><term>Observateur</term>
<term>Robotique</term>
<term>Homme</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">In the majority of robotic and haptic applications, including manipulation and human-robot interaction, contact force needs to be monitored and controlled. Transparent implementation of bilateral teleoperation or haptic controllers necessitates the exchange of operator and environment contact forces. This requires the use of expensive commercially available force/torque sensors, which are rather bulky, are vulnerable to impact forces, and increase system inertia and compliance. An alternative solution is the use of dynamic force observers, which estimate external forces using system dynamic model. However, due to the uncertainties in system dynamic structure and parameters, these model-based observers do not produce accurate force estimates, and often create a dynamic lag that may cause bandwidth limitation and instability. This paper proposes two neural-network-based force/torque observers that do not require a system dynamic model. The observers can estimate human hand force and environment contact force with up to 98.3% accuracy in the sense of mean-square error, and with negligible dynamic lag. The performance of the proposed observers are extensively analyzed in separate human-robot and robot-environment experimental settings, and in a two-channel bilateral teleoperation control loop with multiple runs with two Planar Twin-Pantograph haptic devices.</div>
</front>
</TEI>
<affiliations><list><country><li>Canada</li>
</country>
</list>
<tree><country name="Canada"><noRegion><name sortKey="Smith, Andrew C" sort="Smith, Andrew C" uniqKey="Smith A" first="Andrew C." last="Smith">Andrew C. Smith</name>
</noRegion>
<name sortKey="Hashtrudi Zaad, Keyvan" sort="Hashtrudi Zaad, Keyvan" uniqKey="Hashtrudi Zaad K" first="Keyvan" last="Hashtrudi-Zaad">Keyvan Hashtrudi-Zaad</name>
<name sortKey="Mobasser, Farid" sort="Mobasser, Farid" uniqKey="Mobasser F" first="Farid" last="Mobasser">Farid Mobasser</name>
</country>
</tree>
</affiliations>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/HapticV1/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 006061 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 006061 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Ticri/CIDE |area= HapticV1 |flux= Main |étape= Exploration |type= RBID |clé= Pascal:07-0201416 |texte= Neural-network-based contact force observers for haptic applications }}
![]() | This area was generated with Dilib version V0.6.23. | ![]() |