Improving operational space control of heavy manipulators via open-loop compensation. Maeda, G., Singh, S., & Rye, D. In Proceedings of the International Conference on Intelligent Robots and Systems (IROS), pages 725–731, 2011. IEEE. doi abstract bibtex Disturbances that arise in material removal by repeated attempts to track the same path have the particular characteristics of non-repetitive magnitudes, but nearlyrepetitive or gradual gradient transitions. This paper proposes and validates Iterative Learning Control (ILC) with a PD-type learning function for this class of disturbance as a predictive controller for autonomous excavation. However, parameters of the PD learning function may require different tunings for different excavation conditions, and convergence can be slow when compared to changes in excavation dynamics. In order to improve convergence, a plant inversion learning function is reinterpreted as a disturbance observer in the iteration domain, effectively rendering a disturbance learning controller (DLC). A hydraulic mini-excavator was used to evaluate experimentally the performance of the conventional ILC and the DLC against a robust controller. ILC achieved a desired cut profile with nonmonotonic transients and DLC converged faster by learning disturbances directly from command discrepancies.
@INPROCEEDINGS{maeda2011improving,
author = {Maeda, G.J. and Singh, S.P.N. and Rye, D.C.},
title = {Improving operational space control of heavy manipulators via open-loop
compensation},
booktitle = {Proceedings of the International Conference on Intelligent Robots
and Systems ({IROS})},
year = {2011},
pages = {725--731},
organization = {IEEE},
abstract = {Disturbances that arise in material removal by repeated attempts to
track the same path have the particular characteristics of non-repetitive
magnitudes, but nearlyrepetitive or gradual gradient transitions.
This paper proposes and validates Iterative Learning Control (ILC)
with a PD-type learning function for this class of disturbance as
a predictive controller for autonomous excavation. However, parameters
of the PD learning function may require different tunings for different
excavation conditions, and convergence can be slow when compared
to changes in excavation dynamics. In order to improve convergence,
a plant inversion learning function is reinterpreted as a disturbance
observer in the iteration domain, effectively rendering a disturbance
learning controller (DLC). A hydraulic mini-excavator was used to
evaluate experimentally the performance of the conventional ILC and
the DLC against a robust controller. ILC achieved a desired cut profile
with nonmonotonic transients and DLC converged faster by learning
disturbances directly from command discrepancies.},
doi = {10.1109/IROS.2011.6094437},
pdf = {Maeda2012_IROS_LearningDisturbances.pdf},
review = {Disturbances that arise in material removal by repeated attempts to
track the same path have the particular characteristics of non-repetitive
magnitudes, but nearly-repetitive or gradual gradient transitions.
This paper proposes and validates Iterative Learning Control (ILC)
for this class of disturbance as a low-level control strategy in
autonomous excavation. Parameters of the ILC learning rule may require
different tunings for different excavation conditions, and convergence
can be slow when compared to changes in excavation dynamics. An iterative
Disturbance Learning Controller (DLC) based on inverse dynamics is
presented as a solution to those problems. The DLC learning rule
is independent of the excavation condition and the use of an inverse
arm dynamics accelerates convergence. A hydraulic mini-excavator
was used to evaluate experimentally the performance of the ILC and
the DLC against a robust controller. ILC achieved a desired cut profile
with non-monotonic transients and DLC converged faster by learning
disturbances directly from command discrepancies.}
}
Downloads: 0
{"_id":{"_str":"51fd4b4ac5b22c38760016ee"},"__v":34,"authorIDs":["25JLGLMgHDQdFcbEd","2pNN7j4TWs2sdPq2b","34dFKNX3cFW2RkuZi","35Xv7osbm3zXk62F8","3SKHrAB8wMRysGFy8","47GNnFb6EPKma3Srt","48H54WhqNYjnD95ui","4SntEAiXntKARkNgS","5462d4748a9aab071c0005c7","5T75juAmsAB4z6bD6","5WvQsNxqBRyW2RSQi","5e34214741f782de01000022","5e3426cb41f782de01000084","5e3429b441f782de010000b9","5e34312341f782de0100013e","5e34390b41f782de010001e6","5e34392441f782de010001eb","5e6b0e1e86bf9cde0100000e","5sfq2WJ4W5g8mskFe","6a8JHmpJfRqSEQkqb","6hGsP8HLkwinYXigD","6zbapqaBfYNicQRXv","7CkZJwcgcj75uyjun","7DA6BoSFi7tnffK8H","7WRSvEmwiLfubqe8f","8HnjdunXqJG4QDLAd","8ntKbqqhqJyvrfK6r","9zqTiopbnCKBBJEHw","B7m7NjLbfonZHkc6g","BifZh9Ad6N9fnehRi","CL6tPXijZ3hpfr78G","CmEkRJvYX5Tr4Pnfk","DExNFWbRJaceGfGum","DY3kZm8P2JayQgduy","EyvXqoGdM2qS93CN7","Fdd27HCHQMmQXfGLs","GqK42ji4y7PtPXZ3F","Gy3fvTPGQkczs42ez","HLcHzprYn3r3Xe84b","HdmdEWKvJektnYR7y","Hj7KYhXTskqMzhg7u","HzDPAyHXF2wTA7zGu","JHBPkDMYpmfwpcT8A","JMwRZPisoDhWDzWhm","K37JaQLteDSHdqyjx","K7AXubjaKLhA3pNTA","KqPDgQxE6zxpxX67e","LRjJSNBWHuWanhqNM","LjXJLGxM3QELAoaNP","Mc9S7iNhTFJWsRkeb","Mww7bu4yZkujgW5ns","NAGpXYkKmHQ29Mvjb","NKMyNZac6Y89jH6Qr","PZGS54s4e74ZKchZy","PcR3YBj9m6B3nvjG4","Q7eQ2ycHNYkfAF6nP","QR4SjtXbj8siaJg4K","QReRjpmwvPkeefcfK","QkREvvTHfouEvisjF","Rg9JW68PMYmedw2jv","RxHhvDjoid9t5F7uJ","Sju6B3BiCKw3CtwpM","TYZYfD6XShucnrAkf","XviLSraASaqofFM4Z","XwzuwxPG3rMWe5q7S","XzKRCGm3o924pvPcu","YHiJHk6hwmzPyR4nL","ZezE89A6WSniSjNvd","ZtbB3GZNz96JBnnL6","aFQJQukES2Lw58sS5","aPafrdEuKAitYSfiA","afAXjw75m99Km4Fgj","afwmjw9TTkZfveNzE","bNN2E3KLq5WFdzKTF","cfwHoA5cnQ6eGD4bg","dDWF4crzWnW8RyG9p","dFqdXn6WnDTzTnEqQ","dLkDhYcDM2PSSEZk3","dmJYK8FNXWAAu5G7T","dvJ37AfDNYzAuD9PK","eiYowsdjQ9Mcw2Thz","evxzRHu7G7fbeoibY","fid77jGJy9yzcHJr2","fuEkxmax5jFbtxCYy","gAmxy4SvWarbPBeTX","gFrSDrvd4JrFP9kTP","gXJZBJPrZddgZ7YoH","ggsiWBRuBugZgRE8f","gqQQ75hZq9ros28RR","h3u9YL2Ae2DNrQCqi","hiXcvvFucmwxJEZ9y","hth6XiDc9jGYzRe5H","i2DexC5F5rxsC5PHg","iELGk3WWqbGkdYzPc","ijX6hDPyFBq7xgoma","ip7PgT6GfbdZreojv","is7DZGXrEJDXNvotf","j6o4kAC79ZjvEeijw","jJbSbk2LoomLaPG7h","kTAMCAkfLFo4f89Gp","kXmyRRzRyEi5cz83e","kz4FwssxjbAeByCEF","nRtwjECYR2WAxoeFJ","oDxo3ZMAB4phpqr5o","p64zTgE7XqWDjvB7v","pHCFNdiRdm4dHD4JE","pJd6uJtiZnuhNGMpH","paBtjfzxNGmPkJgFy","pu6XsJacuRdqyidu3","qLFpKhtC29Rk9jeS8","rBLmXAceeFPWAntch","sHdkaG8cdJP4GriPy","tsgNGCAcyhTm8BeQB","u3sfcQ8RovYYpKjLb","uGrPMtPgNJq2bBX47","uHkfdT8WqEWJyDjRg","ucoBD8M4z8ezYE8bc","upFqY33vPBba6wxh8","v9z9AicS8PCAymgiR","vAQtv8TCRg4oNqGSn","vDknP8BNi6WY35SDp","vQekzHzAimvgtFT3G","vg9Go2zuYX2Xq8HJf","vve7zKuHuKZmDZ6FB","w4erytKzHZZj3bnbZ","wmuqnoGDizWJe8P2H","xGz64z5uDaXxM3ZHF","xSFFddhsMR36EEcYu","y52p2nvza87XinHws","yBp5Pq56pbsjD6f8H","z6nH83SNs79e3uH2u","zgA7uYAasRk8ezQ9e","zj2qr55FtWLxKhX9E","zsFBk8a7A7dEouo2u"],"author_short":["Maeda, G.","Singh, S.","Rye, D."],"bibbaseid":"maeda-singh-rye-improvingoperationalspacecontrolofheavymanipulatorsviaopenloopcompensation-2011","bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"propositions":[],"lastnames":["Maeda"],"firstnames":["G.J."],"suffixes":[]},{"propositions":[],"lastnames":["Singh"],"firstnames":["S.P.N."],"suffixes":[]},{"propositions":[],"lastnames":["Rye"],"firstnames":["D.C."],"suffixes":[]}],"title":"Improving operational space control of heavy manipulators via open-loop compensation","booktitle":"Proceedings of the International Conference on Intelligent Robots and Systems (IROS)","year":"2011","pages":"725–731","organization":"IEEE","abstract":"Disturbances that arise in material removal by repeated attempts to track the same path have the particular characteristics of non-repetitive magnitudes, but nearlyrepetitive or gradual gradient transitions. This paper proposes and validates Iterative Learning Control (ILC) with a PD-type learning function for this class of disturbance as a predictive controller for autonomous excavation. However, parameters of the PD learning function may require different tunings for different excavation conditions, and convergence can be slow when compared to changes in excavation dynamics. In order to improve convergence, a plant inversion learning function is reinterpreted as a disturbance observer in the iteration domain, effectively rendering a disturbance learning controller (DLC). A hydraulic mini-excavator was used to evaluate experimentally the performance of the conventional ILC and the DLC against a robust controller. ILC achieved a desired cut profile with nonmonotonic transients and DLC converged faster by learning disturbances directly from command discrepancies.","doi":"10.1109/IROS.2011.6094437","pdf":"Maeda2012_IROS_LearningDisturbances.pdf","review":"Disturbances that arise in material removal by repeated attempts to track the same path have the particular characteristics of non-repetitive magnitudes, but nearly-repetitive or gradual gradient transitions. This paper proposes and validates Iterative Learning Control (ILC) for this class of disturbance as a low-level control strategy in autonomous excavation. Parameters of the ILC learning rule may require different tunings for different excavation conditions, and convergence can be slow when compared to changes in excavation dynamics. An iterative Disturbance Learning Controller (DLC) based on inverse dynamics is presented as a solution to those problems. The DLC learning rule is independent of the excavation condition and the use of an inverse arm dynamics accelerates convergence. A hydraulic mini-excavator was used to evaluate experimentally the performance of the ILC and the DLC against a robust controller. ILC achieved a desired cut profile with non-monotonic transients and DLC converged faster by learning disturbances directly from command discrepancies.","bibtex":"@INPROCEEDINGS{maeda2011improving,\r\n author = {Maeda, G.J. and Singh, S.P.N. and Rye, D.C.},\r\n title = {Improving operational space control of heavy manipulators via open-loop\r\n\tcompensation},\r\n booktitle = {Proceedings of the International Conference on Intelligent Robots\r\n\tand Systems ({IROS})},\r\n year = {2011},\r\n pages = {725--731},\r\n organization = {IEEE},\r\n abstract = {Disturbances that arise in material removal by repeated attempts to\r\n\ttrack the same path have the particular characteristics of non-repetitive\r\n\tmagnitudes, but nearlyrepetitive or gradual gradient transitions.\r\n\tThis paper proposes and validates Iterative Learning Control (ILC)\r\n\twith a PD-type learning function for this class of disturbance as\r\n\ta predictive controller for autonomous excavation. However, parameters\r\n\tof the PD learning function may require different tunings for different\r\n\texcavation conditions, and convergence can be slow when compared\r\n\tto changes in excavation dynamics. In order to improve convergence,\r\n\ta plant inversion learning function is reinterpreted as a disturbance\r\n\tobserver in the iteration domain, effectively rendering a disturbance\r\n\tlearning controller (DLC). A hydraulic mini-excavator was used to\r\n\tevaluate experimentally the performance of the conventional ILC and\r\n\tthe DLC against a robust controller. ILC achieved a desired cut profile\r\n\twith nonmonotonic transients and DLC converged faster by learning\r\n\tdisturbances directly from command discrepancies.},\r\n doi = {10.1109/IROS.2011.6094437},\r\n pdf = {Maeda2012_IROS_LearningDisturbances.pdf},\r\n review = {Disturbances that arise in material removal by repeated attempts to\r\n\ttrack the same path have the particular characteristics of non-repetitive\r\n\tmagnitudes, but nearly-repetitive or gradual gradient transitions.\r\n\tThis paper proposes and validates Iterative Learning Control (ILC)\r\n\tfor this class of disturbance as a low-level control strategy in\r\n\tautonomous excavation. Parameters of the ILC learning rule may require\r\n\tdifferent tunings for different excavation conditions, and convergence\r\n\tcan be slow when compared to changes in excavation dynamics. An iterative\r\n\tDisturbance Learning Controller (DLC) based on inverse dynamics is\r\n\tpresented as a solution to those problems. The DLC learning rule\r\n\tis independent of the excavation condition and the use of an inverse\r\n\tarm dynamics accelerates convergence. A hydraulic mini-excavator\r\n\twas used to evaluate experimentally the performance of the ILC and\r\n\tthe DLC against a robust controller. ILC achieved a desired cut profile\r\n\twith non-monotonic transients and DLC converged faster by learning\r\n\tdisturbances directly from command discrepancies.}\r\n}\r\n\r\n","author_short":["Maeda, G.","Singh, S.","Rye, D."],"key":"maeda2011improving","id":"maeda2011improving","bibbaseid":"maeda-singh-rye-improvingoperationalspacecontrolofheavymanipulatorsviaopenloopcompensation-2011","role":"author","urls":{},"downloads":0,"html":""},"bibtype":"inproceedings","biburl":"http://robotics.itee.uq.edu.au/~spns/pubcache/SpnS_PubList.bib","downloads":0,"keywords":[],"search_terms":["improving","operational","space","control","heavy","manipulators","via","open","loop","compensation","maeda","singh","rye"],"title":"Improving operational space control of heavy manipulators via open-loop compensation","title_words":["improving","operational","space","control","heavy","manipulators","via","open","loop","compensation"],"year":2011,"dataSources":["zNCf6MTxXnpkNN9Zz"]}