Pub Date : 1990-05-23DOI: 10.23919/ACC.1990.4791156
Roy S. Smith, J. Doyle
The paper considers the problem of estimating, from experimental data, real parameters for a model with uncertainty in the form of both additive noise and norm bounded perturbations. Such models frequently arise in robust control theory, and a framework is introduced for the consideration of experimental data in robust control analysis problems. If the analysis tools applied include robust stability tests for real parameter variations (real ¿), then the framework can be used to address the problem of "robust" parameter identification. While the techniques discussed here can quickly become computationally overwhelming when applied to physical systems and real data, the approach introduces a new way of looking at the identification problem and may be helpful in arriving at a more tractable methodology.
{"title":"Towards a Methodology for Robust Parameter Identification","authors":"Roy S. Smith, J. Doyle","doi":"10.23919/ACC.1990.4791156","DOIUrl":"https://doi.org/10.23919/ACC.1990.4791156","url":null,"abstract":"The paper considers the problem of estimating, from experimental data, real parameters for a model with uncertainty in the form of both additive noise and norm bounded perturbations. Such models frequently arise in robust control theory, and a framework is introduced for the consideration of experimental data in robust control analysis problems. If the analysis tools applied include robust stability tests for real parameter variations (real ¿), then the framework can be used to address the problem of \"robust\" parameter identification. While the techniques discussed here can quickly become computationally overwhelming when applied to physical systems and real data, the approach introduces a new way of looking at the identification problem and may be helpful in arriving at a more tractable methodology.","PeriodicalId":307181,"journal":{"name":"1990 American Control Conference","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1990-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126358801","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1990-05-23DOI: 10.23919/ACC.1990.4791049
K. Glitzenstein, J. Hedrick
Production clutch-to-clutch automatic transmissions contain a number of subcomponents difficult to characterize from a traditional modeling stance. Variation inherent to both manufacturing processes and anticipated wear increase the bounds over which a parameter may be expected to range. The most critical of these is observed to be the value associated with the coefficient of friction of the clutch surfaces. Formulated from Lyapunov stability requirements, three adaptive algorithms are proposed to compensate for these variations. The first, a direct adaptive approach, bases parameter behavior on a sliding control law. The second technique incorporates an indirect least squares estimator term into the direct law, forming a composite adaptive result. The third extends the composite principle from the inertia to the torque phases of the shift. It is shown that identification of the friction parameters will occur with eihter of the techniques, and that the rate of convergence and the stability of the parameter improves with the increasing complexity of the technique applied.
{"title":"Adaptive Control of Automotive Transmissions","authors":"K. Glitzenstein, J. Hedrick","doi":"10.23919/ACC.1990.4791049","DOIUrl":"https://doi.org/10.23919/ACC.1990.4791049","url":null,"abstract":"Production clutch-to-clutch automatic transmissions contain a number of subcomponents difficult to characterize from a traditional modeling stance. Variation inherent to both manufacturing processes and anticipated wear increase the bounds over which a parameter may be expected to range. The most critical of these is observed to be the value associated with the coefficient of friction of the clutch surfaces. Formulated from Lyapunov stability requirements, three adaptive algorithms are proposed to compensate for these variations. The first, a direct adaptive approach, bases parameter behavior on a sliding control law. The second technique incorporates an indirect least squares estimator term into the direct law, forming a composite adaptive result. The third extends the composite principle from the inertia to the torque phases of the shift. It is shown that identification of the friction parameters will occur with eihter of the techniques, and that the rate of convergence and the stability of the parameter improves with the increasing complexity of the technique applied.","PeriodicalId":307181,"journal":{"name":"1990 American Control Conference","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1990-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126434691","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1990-05-23DOI: 10.23919/ACC.1990.4791234
C. Atkeson
We report on a preliminary investigation of tasklevel learning, an approach to learning from practice. We have programmed a robot to juggle a single ball in three dimensions by batting it up-wards with a large paddle. The robot uses a real-time binary vision system to track the ball and measure its performance. Task-level learning consists of building a model of performance errors at the task level during practice, and using that model to refine task-level commands. A polynomial surface was fit to the errors in where the ball went after each hit, and this task model is used to refine how the ball is hit. This application of task-level learning dramatically increased the number of consecutive hits the robot could execute before the ball was hit out of range of the paddle. The talk explores memory-based techniques for future implementations of tasklevel learning.
{"title":"Memory-Based Techniques for Task-Level Learning in Robots and Smart Machines","authors":"C. Atkeson","doi":"10.23919/ACC.1990.4791234","DOIUrl":"https://doi.org/10.23919/ACC.1990.4791234","url":null,"abstract":"We report on a preliminary investigation of tasklevel learning, an approach to learning from practice. We have programmed a robot to juggle a single ball in three dimensions by batting it up-wards with a large paddle. The robot uses a real-time binary vision system to track the ball and measure its performance. Task-level learning consists of building a model of performance errors at the task level during practice, and using that model to refine task-level commands. A polynomial surface was fit to the errors in where the ball went after each hit, and this task model is used to refine how the ball is hit. This application of task-level learning dramatically increased the number of consecutive hits the robot could execute before the ball was hit out of range of the paddle. The talk explores memory-based techniques for future implementations of tasklevel learning.","PeriodicalId":307181,"journal":{"name":"1990 American Control Conference","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1990-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126443507","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1990-05-23DOI: 10.23919/ACC.1990.4790822
S. Choi, B. Thompson, M. Gandhi
This paper discusses on the control of a single-link flexible robotic manipulator fabricated from advanced composite laminates. An output feedback controller incorporating two colocated sensors associated with angular position and velocity is designed based on the reduced-order model and implemented in order to investigate the control performances of robotic arms fabricated from aluminum and composite laminates. The control performances are evaluated and compared between two robotic arms through step responses of the system. It is shown in the experimental investigation that the arm fabricated from composite laminates has superior performance characteristics such as faster settling-time, smaller input torque and a superior increment of stability relative to the arm fabricated from aluminum.
{"title":"A Theoretical and Experimental Investigation on the Control of a Single-Link Flexible Robotic Manipulator Fabricated from Composite Laminates","authors":"S. Choi, B. Thompson, M. Gandhi","doi":"10.23919/ACC.1990.4790822","DOIUrl":"https://doi.org/10.23919/ACC.1990.4790822","url":null,"abstract":"This paper discusses on the control of a single-link flexible robotic manipulator fabricated from advanced composite laminates. An output feedback controller incorporating two colocated sensors associated with angular position and velocity is designed based on the reduced-order model and implemented in order to investigate the control performances of robotic arms fabricated from aluminum and composite laminates. The control performances are evaluated and compared between two robotic arms through step responses of the system. It is shown in the experimental investigation that the arm fabricated from composite laminates has superior performance characteristics such as faster settling-time, smaller input torque and a superior increment of stability relative to the arm fabricated from aluminum.","PeriodicalId":307181,"journal":{"name":"1990 American Control Conference","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1990-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128045608","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1990-05-23DOI: 10.23919/ACC.1990.4791037
A. Tornambè
It is shown that the asymptotic inverse kinematic problem can be reduced to the problem of observing the state variables of a certain nonlinear dynamic system. A simple asymptotic observer is used in the state estimation and the singular perturbation theory is used in the convergence proof of the algorithm. It is also shown that the classical methods of Newton and of the gradient can be obtained as particular asymptotic observers.
{"title":"An Asymptotic Observer for Solving the Inverse Kinematic Problem","authors":"A. Tornambè","doi":"10.23919/ACC.1990.4791037","DOIUrl":"https://doi.org/10.23919/ACC.1990.4791037","url":null,"abstract":"It is shown that the asymptotic inverse kinematic problem can be reduced to the problem of observing the state variables of a certain nonlinear dynamic system. A simple asymptotic observer is used in the state estimation and the singular perturbation theory is used in the convergence proof of the algorithm. It is also shown that the classical methods of Newton and of the gradient can be obtained as particular asymptotic observers.","PeriodicalId":307181,"journal":{"name":"1990 American Control Conference","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1990-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128112950","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1990-05-23DOI: 10.23919/ACC.1990.4790930
Mehrez Hached, S. Madani-Esfahani, S. Żak
This paper addresses the problem of stabilization of a class of uncertain and/or nonlinear dynamic systems for which matching conditions are not satisfied. No statistical description of uncertain elements is assumed. The uncertain and/or nonlinear quantities are described only in terms of bounds and their possible sizes. A stabilizing bounded controller is proposed whose design utilizes elements of the variable structure control theory. Regions of uniform ultimate boundedness (practical stability) for systems driven by this bounded controller are estimated. A special coordinate transformation is utilized during the analysis. Sliding domains and estimates of the domain of attraction are obtained with the aid of the Lyapunov method.
{"title":"On the Stability and Estimation of Ultimate Boundedness of Nonlinear/Uncertain Dynamic Systems with Bounded Controllers","authors":"Mehrez Hached, S. Madani-Esfahani, S. Żak","doi":"10.23919/ACC.1990.4790930","DOIUrl":"https://doi.org/10.23919/ACC.1990.4790930","url":null,"abstract":"This paper addresses the problem of stabilization of a class of uncertain and/or nonlinear dynamic systems for which matching conditions are not satisfied. No statistical description of uncertain elements is assumed. The uncertain and/or nonlinear quantities are described only in terms of bounds and their possible sizes. A stabilizing bounded controller is proposed whose design utilizes elements of the variable structure control theory. Regions of uniform ultimate boundedness (practical stability) for systems driven by this bounded controller are estimated. A special coordinate transformation is utilized during the analysis. Sliding domains and estimates of the domain of attraction are obtained with the aid of the Lyapunov method.","PeriodicalId":307181,"journal":{"name":"1990 American Control Conference","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1990-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125719565","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1990-05-23DOI: 10.23919/ACC.1990.4790704
L. Hideg, R. Judd, R. Van Til
The error between the desired and actual output of a single input, multiple output plant is used by a learning law to modify the input in order to reduce the output error. Lyapunov stability analysis is used to determine criteria for the learning law coefficients. The plant is assumed to be stable, linear and time invariant. Formulations include restrictions on the desired trajectory and Lyapunov candidate derived from the plant model. Sufficient conditions for stability are found and conjectured to also be necessary.
{"title":"Stability Analysis for Learning Systems","authors":"L. Hideg, R. Judd, R. Van Til","doi":"10.23919/ACC.1990.4790704","DOIUrl":"https://doi.org/10.23919/ACC.1990.4790704","url":null,"abstract":"The error between the desired and actual output of a single input, multiple output plant is used by a learning law to modify the input in order to reduce the output error. Lyapunov stability analysis is used to determine criteria for the learning law coefficients. The plant is assumed to be stable, linear and time invariant. Formulations include restrictions on the desired trajectory and Lyapunov candidate derived from the plant model. Sufficient conditions for stability are found and conjectured to also be necessary.","PeriodicalId":307181,"journal":{"name":"1990 American Control Conference","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1990-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121962865","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1990-05-23DOI: 10.1109/ACC.1990.4174097
Y. Chait, Neil F. Lacey, C. MacCluer
A new methodology is presented for deriving transient response estimates for a large class of uncertain linear feedback SISO systems, including unstable distributed parameter systems. It is shown that, using operational calculus, estimates exist and can be computed even in cases where a certain error norm is greater than unity. Several simple examples illustrate the application of the method.
{"title":"Using Operational Calculus to Obtain Transient Response Estimates in Linear Uncertain Feedback Systems","authors":"Y. Chait, Neil F. Lacey, C. MacCluer","doi":"10.1109/ACC.1990.4174097","DOIUrl":"https://doi.org/10.1109/ACC.1990.4174097","url":null,"abstract":"A new methodology is presented for deriving transient response estimates for a large class of uncertain linear feedback SISO systems, including unstable distributed parameter systems. It is shown that, using operational calculus, estimates exist and can be computed even in cases where a certain error norm is greater than unity. Several simple examples illustrate the application of the method.","PeriodicalId":307181,"journal":{"name":"1990 American Control Conference","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1990-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127995731","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1990-05-23DOI: 10.1109/ACC.1990.4174264
R. Luus
The difficulty in obtaining the optimal control policy, where singular subarcs appear when Pontryagin's Maximum Principle is applied, does not arise if dynamic programming is used. To apply dynamic programming, a relatively coarse grid is used for the state variables and only a small number of allowable values of control are taken at each time stage. The grid and the range for allowable control are contracted after every iteration to provide an accurate control policy after a sufficient number of iterations. The effects of the coarseness of the grid and the size of the reduction factor are examined with a typical chemical engineering problem that has been found to be difficult to solve by other means. To reduce computational effort for high-dimensional systems, it is recommended to use for the grid elements only accessible states. The choice of such grid elements is examined by considering an example of higher dimension. The proposed procedure is easy to program and yields better results than have been reported in the literature for the systems considered.
{"title":"Application of Dynamic Programming to Singular Optimal Control Problems","authors":"R. Luus","doi":"10.1109/ACC.1990.4174264","DOIUrl":"https://doi.org/10.1109/ACC.1990.4174264","url":null,"abstract":"The difficulty in obtaining the optimal control policy, where singular subarcs appear when Pontryagin's Maximum Principle is applied, does not arise if dynamic programming is used. To apply dynamic programming, a relatively coarse grid is used for the state variables and only a small number of allowable values of control are taken at each time stage. The grid and the range for allowable control are contracted after every iteration to provide an accurate control policy after a sufficient number of iterations. The effects of the coarseness of the grid and the size of the reduction factor are examined with a typical chemical engineering problem that has been found to be difficult to solve by other means. To reduce computational effort for high-dimensional systems, it is recommended to use for the grid elements only accessible states. The choice of such grid elements is examined by considering an example of higher dimension. The proposed procedure is easy to program and yields better results than have been reported in the literature for the systems considered.","PeriodicalId":307181,"journal":{"name":"1990 American Control Conference","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1990-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130139329","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1990-05-23DOI: 10.23919/ACC.1990.4790841
J. Ackermann, W. Sienel
Robust control problems in automatic steering are due to the wide range of velocity, mass and road conditions under which such vehicles operate. In earlier design studies and road tests for a bus it has been shown that it is possible to design a fixed gain controller such that the automatic steering operates satisfactorily over the entire range of parameters. In the present paper it is shown that the performance of such a robust automatic steering system can be considerably improved by the addition of a gyro measuring the yaw rate and feeding it back into the controller.
{"title":"Robust Control for Automatic Steering","authors":"J. Ackermann, W. Sienel","doi":"10.23919/ACC.1990.4790841","DOIUrl":"https://doi.org/10.23919/ACC.1990.4790841","url":null,"abstract":"Robust control problems in automatic steering are due to the wide range of velocity, mass and road conditions under which such vehicles operate. In earlier design studies and road tests for a bus it has been shown that it is possible to design a fixed gain controller such that the automatic steering operates satisfactorily over the entire range of parameters. In the present paper it is shown that the performance of such a robust automatic steering system can be considerably improved by the addition of a gyro measuring the yaw rate and feeding it back into the controller.","PeriodicalId":307181,"journal":{"name":"1990 American Control Conference","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1990-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130186746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：