Proceedings - 2010 Latin American Robotics Symposium and Intelligent Robotics Meeting, LARS 2010, 2010
... Model Predictive Control is usually designed from the discrete-time model of the system. ... ... more ... Model Predictive Control is usually designed from the discrete-time model of the system. ... in order to enforce a fast convergence of state trajectories to ¯x. As commented beforehand, in order to avoid nonlinear constraints, control constraints are imposed only in time step k ...
2008 16th Mediterranean Conference on Control and Automation, 2008
This paper presents a new algorithm for model predictive control (MPC) of constrained bilinear sy... more This paper presents a new algorithm for model predictive control (MPC) of constrained bilinear systems using iterative compensation of the prediction error and invariant sets for constraints satisfaction and stability guarantee. In order to improve the performance of the controller, which holds prediction as its essence, an iterative process is proposed with the objective of reducing the prediction errors due
2011 9th IEEE International Conference on Control and Automation (ICCA), 2011
This paper presents a new algorithm for bilinear predictive control based on state variables. Thi... more This paper presents a new algorithm for bilinear predictive control based on state variables. This algorithm uses a time-step quasilinear model and adds compensations terms to the predictor model, which are different for each prediction horizon. A simulated example shows an improvement in the control performance by using this method, when it is compared with the “time-step quasilinear controller”.
2011 19th Mediterranean Conference on Control & Automation (MED), 2011
Nonlinear Predictive Control has been subject of many researches in recent decades. Bilinear mode... more Nonlinear Predictive Control has been subject of many researches in recent decades. Bilinear models have been an alternative to represent process nonlinearities because they are simpler than the nonlinear models in general and satisfactorily represent many types of nonlinearities. This paper presents a state variables approach of bilinear predictive control. This approach uses a compensated state variables model obtained from the compensated polynomial model, which uses a quasilinear model with the addition of compensations terms to the predictor model. These terms are different for each prediction horizon. A simulated example shows an improvement in the control performance by using this method, when it is compared with the "time-step quasilinear predictive controller".
ABSTRACT This paper presents and discusses the implementation results of a model-predictive contr... more ABSTRACT This paper presents and discusses the implementation results of a model-predictive control (MPC) scheme with friction compensation applied to trajectory following of an omnidirectional three-wheeled robot. A cascade structure is used with an inverse kinematics block to generate the velocity references given to the predictive controller. Part of the control effort is used to compensate for the effects of static friction, allowing the use of efficient algorithms for linear MPC with constraints. Experimental results show that the proposed strategy is efficient in compensating for frictional effects as well as for tracking predefined trajectories.
This paper proposes an extension of some results on the l1-disturbance attenuation problem based ... more This paper proposes an extension of some results on the l1-disturbance attenuation problem based on the approach by D-(A,B)-invariance of polyhedral domains. This extension concerns the case of discrete controllers, for which, at each period, the control vector can only take a finite number of values. An example of such limitations is the respect of lot-sizes in production activities. This example is described and a candidate discrete controller is proposed for solving in closed-loop multistage planning problems.
This paper aims to present the use of Neural Networks (NN) to approximate the input-output map ma... more This paper aims to present the use of Neural Networks (NN) to approximate the input-output map manifold for online output injection laws in the context of observers with error limitation. The set invariance approach is applied for computation of as small as possible conditioned invariant sets that confine the estimation error of a full-order observer. Online output injection laws are attractive due to its ability to cope with the peak phenomenon, but the runtime of its computation may turn its use unfeasible in systems with fast dynamics. The input-output map of such injection law is then approximated via NN training. Structures with NNs present processing times as fast as the adequate technology is employed for their implementation, for instance, FPGAs. Two examples, one of which the classical well-known magnetic levitation ball system, are presented to illustrate the merit of the approach. Resumo— Neste artigo, o uso de Redes Neurais Artificiais para a aproximação do mapa da varie...
2006 49th IEEE International Midwest Symposium on Circuits and Systems, 2006
This work presents an automatic sizing tool for optimal design of CMOS OTA, where the formulation... more This work presents an automatic sizing tool for optimal design of CMOS OTA, where the formulation is based on the Advanced Compact MOSFET (ACM) model. Several design examples prove that the methodology is reliable, provides useful guidelines, exploits the entire design space and is also adequate for small load capacitances.
2013 Latin American Robotics Symposium and Competition, 2013
ABSTRACT Concerning the human factor and ergonomics in an active orthosis for lower limbs applica... more ABSTRACT Concerning the human factor and ergonomics in an active orthosis for lower limbs application, it is important to generate an anthropomorphic gait. The objective of this work is to find an anthropomorphic gait model using Principal Component Analysis able to be used in a real orthosis application. The gait was modeled from the perspective of the joint leg angles. In order to find a model close to the human gait behavior, gait experiments from humans were collected. The straight line slow walk was selected to be modeled because it fits the basic walk for an orthosis. The validity of the proposed model was verified through extrapolation and comparison with real human gait cycles, different from the ones used to build the model.
This paper aims to present an approach for design of dynamic output feedback compensators for lin... more This paper aims to present an approach for design of dynamic output feedback compensators for linear discrete-time descriptor systems subject to state and control constraints. To this end, output feedback controlled invariant polyhedra are constructed
by taking a pair of polyhedral sets: a controlled invariant set and a conditioned invariant set. By defining an augmented system composed of the original system plus the dynamic compensator, a control action can be computed online, which optimizes the contraction rate of the augmented state trajectory and enforces the constraints. The results are illustrated through numerical examples, which show that the proposed dynamic compensators
outperform static feedback controllers under the same conditions.
A novel technique is proposed for output feedback control of constrained linear discrete-time cau... more A novel technique is proposed for output feedback control of constrained linear discrete-time causal descriptor systems. In particular, it is shown how linear state and control constraints can be satisfied by using only output measurements corrupted by unknown but bounded noise. To this end, conditions are established under which a polyhedron contained in the set defined by the state constraints can be made invariant through output feedback. The control input which enforces the constraints can be computed online through the solution of linear programming problems. The proposed technique is then applied to the control of a hydraulic three-tank system. Simulated as well as experimental results illustrate the proposed technique and indicate that it can be potentially used in practical situations.
A novel descriptor observer structure is proposed in the context of discrete-time descriptor line... more A novel descriptor observer structure is proposed in the context of discrete-time descriptor linear systems. On the basis of this new structure, the characterization of conditioned-invariant polyhedral sets with the purpose of limiting the estimation error is carried out. Limitation of the estimation error can be achieved by the computation of an as small as possible conditioned-invariant polyhedron, that contains the set of possible initial errors, and by a suitable output injection law. The approach is applied to an experimental tanks platform, for which a descriptor model is identified and validated. The obtained results confirm the merits of the proposed observer for applications in this kind of systems.
This work adresses the Disturbance Decoupling Problem in linear systems via static output feedbac... more This work adresses the Disturbance Decoupling Problem in linear systems via static output feedback. Necessary and sufficient conditions for solvability in two important families of systems are established. The problem is solvable if and only if a given subspace verifies an invariance property. The set of output feedback matrices which solve the problem is then parameterized through a suitable change of the coordinate basis of the state, input and output spaces. A numerical example illustrates the proposed approach.
The disturbance decoupling problem (DDP) is solved for a class of linear systems where the left-i... more The disturbance decoupling problem (DDP) is solved for a class of linear systems where the left-invertible systems are included. The authors obtain a complete characterization of the solution, as well as the parameterization of all corresponding state feedback matrices. Such characterization enables one to conclude about the solvability of DDP with stability and/or pole placement. A numerical method is proposed for computation of the controllers in an analytical form, suitable to the treatment of complementary design specifications
Proceedings - 2010 Latin American Robotics Symposium and Intelligent Robotics Meeting, LARS 2010, 2010
... Model Predictive Control is usually designed from the discrete-time model of the system. ... ... more ... Model Predictive Control is usually designed from the discrete-time model of the system. ... in order to enforce a fast convergence of state trajectories to ¯x. As commented beforehand, in order to avoid nonlinear constraints, control constraints are imposed only in time step k ...
2008 16th Mediterranean Conference on Control and Automation, 2008
This paper presents a new algorithm for model predictive control (MPC) of constrained bilinear sy... more This paper presents a new algorithm for model predictive control (MPC) of constrained bilinear systems using iterative compensation of the prediction error and invariant sets for constraints satisfaction and stability guarantee. In order to improve the performance of the controller, which holds prediction as its essence, an iterative process is proposed with the objective of reducing the prediction errors due
2011 9th IEEE International Conference on Control and Automation (ICCA), 2011
This paper presents a new algorithm for bilinear predictive control based on state variables. Thi... more This paper presents a new algorithm for bilinear predictive control based on state variables. This algorithm uses a time-step quasilinear model and adds compensations terms to the predictor model, which are different for each prediction horizon. A simulated example shows an improvement in the control performance by using this method, when it is compared with the “time-step quasilinear controller”.
2011 19th Mediterranean Conference on Control & Automation (MED), 2011
Nonlinear Predictive Control has been subject of many researches in recent decades. Bilinear mode... more Nonlinear Predictive Control has been subject of many researches in recent decades. Bilinear models have been an alternative to represent process nonlinearities because they are simpler than the nonlinear models in general and satisfactorily represent many types of nonlinearities. This paper presents a state variables approach of bilinear predictive control. This approach uses a compensated state variables model obtained from the compensated polynomial model, which uses a quasilinear model with the addition of compensations terms to the predictor model. These terms are different for each prediction horizon. A simulated example shows an improvement in the control performance by using this method, when it is compared with the "time-step quasilinear predictive controller".
ABSTRACT This paper presents and discusses the implementation results of a model-predictive contr... more ABSTRACT This paper presents and discusses the implementation results of a model-predictive control (MPC) scheme with friction compensation applied to trajectory following of an omnidirectional three-wheeled robot. A cascade structure is used with an inverse kinematics block to generate the velocity references given to the predictive controller. Part of the control effort is used to compensate for the effects of static friction, allowing the use of efficient algorithms for linear MPC with constraints. Experimental results show that the proposed strategy is efficient in compensating for frictional effects as well as for tracking predefined trajectories.
This paper proposes an extension of some results on the l1-disturbance attenuation problem based ... more This paper proposes an extension of some results on the l1-disturbance attenuation problem based on the approach by D-(A,B)-invariance of polyhedral domains. This extension concerns the case of discrete controllers, for which, at each period, the control vector can only take a finite number of values. An example of such limitations is the respect of lot-sizes in production activities. This example is described and a candidate discrete controller is proposed for solving in closed-loop multistage planning problems.
This paper aims to present the use of Neural Networks (NN) to approximate the input-output map ma... more This paper aims to present the use of Neural Networks (NN) to approximate the input-output map manifold for online output injection laws in the context of observers with error limitation. The set invariance approach is applied for computation of as small as possible conditioned invariant sets that confine the estimation error of a full-order observer. Online output injection laws are attractive due to its ability to cope with the peak phenomenon, but the runtime of its computation may turn its use unfeasible in systems with fast dynamics. The input-output map of such injection law is then approximated via NN training. Structures with NNs present processing times as fast as the adequate technology is employed for their implementation, for instance, FPGAs. Two examples, one of which the classical well-known magnetic levitation ball system, are presented to illustrate the merit of the approach. Resumo— Neste artigo, o uso de Redes Neurais Artificiais para a aproximação do mapa da varie...
2006 49th IEEE International Midwest Symposium on Circuits and Systems, 2006
This work presents an automatic sizing tool for optimal design of CMOS OTA, where the formulation... more This work presents an automatic sizing tool for optimal design of CMOS OTA, where the formulation is based on the Advanced Compact MOSFET (ACM) model. Several design examples prove that the methodology is reliable, provides useful guidelines, exploits the entire design space and is also adequate for small load capacitances.
2013 Latin American Robotics Symposium and Competition, 2013
ABSTRACT Concerning the human factor and ergonomics in an active orthosis for lower limbs applica... more ABSTRACT Concerning the human factor and ergonomics in an active orthosis for lower limbs application, it is important to generate an anthropomorphic gait. The objective of this work is to find an anthropomorphic gait model using Principal Component Analysis able to be used in a real orthosis application. The gait was modeled from the perspective of the joint leg angles. In order to find a model close to the human gait behavior, gait experiments from humans were collected. The straight line slow walk was selected to be modeled because it fits the basic walk for an orthosis. The validity of the proposed model was verified through extrapolation and comparison with real human gait cycles, different from the ones used to build the model.
This paper aims to present an approach for design of dynamic output feedback compensators for lin... more This paper aims to present an approach for design of dynamic output feedback compensators for linear discrete-time descriptor systems subject to state and control constraints. To this end, output feedback controlled invariant polyhedra are constructed
by taking a pair of polyhedral sets: a controlled invariant set and a conditioned invariant set. By defining an augmented system composed of the original system plus the dynamic compensator, a control action can be computed online, which optimizes the contraction rate of the augmented state trajectory and enforces the constraints. The results are illustrated through numerical examples, which show that the proposed dynamic compensators
outperform static feedback controllers under the same conditions.
A novel technique is proposed for output feedback control of constrained linear discrete-time cau... more A novel technique is proposed for output feedback control of constrained linear discrete-time causal descriptor systems. In particular, it is shown how linear state and control constraints can be satisfied by using only output measurements corrupted by unknown but bounded noise. To this end, conditions are established under which a polyhedron contained in the set defined by the state constraints can be made invariant through output feedback. The control input which enforces the constraints can be computed online through the solution of linear programming problems. The proposed technique is then applied to the control of a hydraulic three-tank system. Simulated as well as experimental results illustrate the proposed technique and indicate that it can be potentially used in practical situations.
A novel descriptor observer structure is proposed in the context of discrete-time descriptor line... more A novel descriptor observer structure is proposed in the context of discrete-time descriptor linear systems. On the basis of this new structure, the characterization of conditioned-invariant polyhedral sets with the purpose of limiting the estimation error is carried out. Limitation of the estimation error can be achieved by the computation of an as small as possible conditioned-invariant polyhedron, that contains the set of possible initial errors, and by a suitable output injection law. The approach is applied to an experimental tanks platform, for which a descriptor model is identified and validated. The obtained results confirm the merits of the proposed observer for applications in this kind of systems.
This work adresses the Disturbance Decoupling Problem in linear systems via static output feedbac... more This work adresses the Disturbance Decoupling Problem in linear systems via static output feedback. Necessary and sufficient conditions for solvability in two important families of systems are established. The problem is solvable if and only if a given subspace verifies an invariance property. The set of output feedback matrices which solve the problem is then parameterized through a suitable change of the coordinate basis of the state, input and output spaces. A numerical example illustrates the proposed approach.
The disturbance decoupling problem (DDP) is solved for a class of linear systems where the left-i... more The disturbance decoupling problem (DDP) is solved for a class of linear systems where the left-invertible systems are included. The authors obtain a complete characterization of the solution, as well as the parameterization of all corresponding state feedback matrices. Such characterization enables one to conclude about the solvability of DDP with stability and/or pole placement. A numerical method is proposed for computation of the controllers in an analytical form, suitable to the treatment of complementary design specifications
Uploads
Papers by Carlos Dorea
by taking a pair of polyhedral sets: a controlled invariant set and a conditioned invariant set. By defining an augmented system composed of the original system plus the dynamic compensator, a control action can be computed online, which optimizes the contraction rate of the augmented state trajectory and enforces the constraints. The results are illustrated through numerical examples, which show that the proposed dynamic compensators
outperform static feedback controllers under the same conditions.
by taking a pair of polyhedral sets: a controlled invariant set and a conditioned invariant set. By defining an augmented system composed of the original system plus the dynamic compensator, a control action can be computed online, which optimizes the contraction rate of the augmented state trajectory and enforces the constraints. The results are illustrated through numerical examples, which show that the proposed dynamic compensators
outperform static feedback controllers under the same conditions.