The Experts below are selected from a list of 50967 Experts worldwide ranked by ideXlab platform
Tyler H. Summers - One of the best experts on this subject based on the ideXlab platform.
-
IROS - Minimax Iterative Dynamic Game: Application to Nonlinear Robot Control Tasks
2018 IEEE RSJ International Conference on Intelligent Robots and Systems (IROS), 2018Co-Authors: Olalekan Ogunmolu, Nicholas Gans, Tyler H. SummersAbstract:Multistage decision policies provide useful control strategies in high-dimensional state spaces, particularly in complex control tasks. However, they exhibit weak performance guarantees in the presence of disturbance, model mismatch, or model uncertainties. This brittleness limits their use in high-risk scenarios. We present how to quantify the sensitivity of such policies in order to inform of their robustness capacity. We also propose a minimax iterative Dynamic Game framework for designing robust policies in the presence of disturbance/uncertainties. We test the quantification hypothesis on a carefully designed deep neural network policy; we then pose a minimax iterative Dynamic Game (iDG) framework for improving policy robustness in the presence of adversarial disturbances. We evaluate our iDG framework on a mecanum-wheeled robot, whose goal is to find a ocally robust optimal multistage policy that achieve a given goal-reaching task. The algorithm is simple and adaptable for designing meta-learning/deep policies that are robust against disturbances, model mismatch, or model uncertainties, up to a disturbance bound. Videos of the results are on the author's website: https://goo.gl/JhshTB, while the codes for reproducing our experiments are on github: https://goo.gl/3G2VBy. A self-contained environment for reproducing our results is on docker: https://goo.gllB07MB $\text{e}$.
-
Minimax Iterative Dynamic Game: Application to Nonlinear Robot Control Tasks
arXiv: Robotics, 2017Co-Authors: Olalekan Ogunmolu, Nicholas Gans, Tyler H. SummersAbstract:Multistage decision policies provide useful control strategies in high-dimensional state spaces, particularly in complex control tasks. However, they exhibit weak performance guarantees in the presence of disturbance, model mismatch, or model uncertainties. This brittleness limits their use in high-risk scenarios. We present how to quantify the sensitivity of such policies in order to inform of their robustness capacity. We also propose a minimax iterative Dynamic Game framework for designing robust policies in the presence of disturbance/uncertainties. We test the quantification hypothesis on a carefully designed deep neural network policy; we then pose a minimax iterative Dynamic Game (iDG) framework for improving policy robustness in the presence of adversarial disturbances. We evaluate our iDG framework on a mecanum-wheeled robot, whose goal is to find a ocally robust optimal multistage policy that achieve a given goal-reaching task. The algorithm is simple and adaptable for designing meta-learning/deep policies that are robust against disturbances, model mismatch, or model uncertainties, up to a disturbance bound. Videos of the results are on the author's website, this http URL, while the codes for reproducing our experiments are on github, this https URL. A self-contained environment for reproducing our results is on docker, this https URL
Ian R. Petersen - One of the best experts on this subject based on the ideXlab platform.
-
Finite Horizon H∞ Control for a Class of Linear Quantum Sampled-Data Measurement Systems: A Dynamic Game Approach*
IFAC Proceedings Volumes, 2011Co-Authors: Aline I. Maalouf, Ian R. PetersenAbstract:Abstract In this paper, the finite horizon H∞ control problem is solved for a class of linear quantum systems using a Dynamic Game approach for the case of sampled-data measurements. The methodology adopted involves a certain equivalence between the quantum problem and an auxiliary classical stochastic problem. Then, by solving the finite horizon H∞ control problem for the equivalent stochastic problem using some results from a corresponding deterministic problem following a Dynamic Game approach, the finite horizon H∞ control problem for the class of linear quantum systems under consideration is solved for the case of sampled-data measurements.
-
ACC - Finite horizon H ∞ control for a class of linear quantum measurement delayed systems: A Dynamic Game approach
Proceedings of the 2011 American Control Conference, 2011Co-Authors: Aline I. Maalouf, Ian R. PetersenAbstract:In this paper, a finite horizon H∞ control problem is solved for a class of linear quantum systems using a Dynamic Game approach for the case of delayed measurements. The methodology adopted involves an equivalence between the quantum problem and an auxiliary classical stochastic problem. Then, the finite horizon H∞ control problem for the class of linear quantum systems under consideration is solved for the case of delayed measurements by solving the finite horizon H∞ control problem for an equivalent stochastic H∞ control problem using some results from a corresponding deterministic problem following a Dynamic Game approach.
-
Finite horizon H ∞ control for a class of linear quantum systems: A Dynamic Game approach
Proceedings of the 2010 American Control Conference, 2010Co-Authors: Aline I. Maalouf, Ian R. PetersenAbstract:In this paper, the finite horizon H∞ control problem is solved for a class of linear quantum systems using a Dynamic Game approach. The methodology adopted involves an equivalence between the quantum problem and an auxiliary classical stochastic problem. Then, by solving the finite horizon H∞ control problem for the equivalent stochastic problem using results from a corresponding deterministic problem following a Dynamic Game approach, the finite horizon H∞ control problem for the class of linear quantum systems under consideration is solved.
Olalekan Ogunmolu - One of the best experts on this subject based on the ideXlab platform.
-
IROS - Minimax Iterative Dynamic Game: Application to Nonlinear Robot Control Tasks
2018 IEEE RSJ International Conference on Intelligent Robots and Systems (IROS), 2018Co-Authors: Olalekan Ogunmolu, Nicholas Gans, Tyler H. SummersAbstract:Multistage decision policies provide useful control strategies in high-dimensional state spaces, particularly in complex control tasks. However, they exhibit weak performance guarantees in the presence of disturbance, model mismatch, or model uncertainties. This brittleness limits their use in high-risk scenarios. We present how to quantify the sensitivity of such policies in order to inform of their robustness capacity. We also propose a minimax iterative Dynamic Game framework for designing robust policies in the presence of disturbance/uncertainties. We test the quantification hypothesis on a carefully designed deep neural network policy; we then pose a minimax iterative Dynamic Game (iDG) framework for improving policy robustness in the presence of adversarial disturbances. We evaluate our iDG framework on a mecanum-wheeled robot, whose goal is to find a ocally robust optimal multistage policy that achieve a given goal-reaching task. The algorithm is simple and adaptable for designing meta-learning/deep policies that are robust against disturbances, model mismatch, or model uncertainties, up to a disturbance bound. Videos of the results are on the author's website: https://goo.gl/JhshTB, while the codes for reproducing our experiments are on github: https://goo.gl/3G2VBy. A self-contained environment for reproducing our results is on docker: https://goo.gllB07MB $\text{e}$.
-
Minimax Iterative Dynamic Game: Application to Nonlinear Robot Control Tasks
arXiv: Robotics, 2017Co-Authors: Olalekan Ogunmolu, Nicholas Gans, Tyler H. SummersAbstract:Multistage decision policies provide useful control strategies in high-dimensional state spaces, particularly in complex control tasks. However, they exhibit weak performance guarantees in the presence of disturbance, model mismatch, or model uncertainties. This brittleness limits their use in high-risk scenarios. We present how to quantify the sensitivity of such policies in order to inform of their robustness capacity. We also propose a minimax iterative Dynamic Game framework for designing robust policies in the presence of disturbance/uncertainties. We test the quantification hypothesis on a carefully designed deep neural network policy; we then pose a minimax iterative Dynamic Game (iDG) framework for improving policy robustness in the presence of adversarial disturbances. We evaluate our iDG framework on a mecanum-wheeled robot, whose goal is to find a ocally robust optimal multistage policy that achieve a given goal-reaching task. The algorithm is simple and adaptable for designing meta-learning/deep policies that are robust against disturbances, model mismatch, or model uncertainties, up to a disturbance bound. Videos of the results are on the author's website, this http URL, while the codes for reproducing our experiments are on github, this https URL. A self-contained environment for reproducing our results is on docker, this https URL
Guiomar Martinherran - One of the best experts on this subject based on the ideXlab platform.
-
spatial effects and strategic behavior in a multiregional transboundary pollution Dynamic Game
Journal of Environmental Economics and Management, 2017Co-Authors: Javier De Frutos, Guiomar MartinherranAbstract:We analyze a transboundary pollution differential Game where pollution control is spatially distributed among a number of agents with predetermined spatial relationships. The analysis emphasizes, first, the effects of the different geographical relationships among decision makers; and second, the strategic behaviour of the agents. The Dynamic Game considers a pollution stock (the state variable) distributed among one large region divided in subregions which control their own emissions of pollutants. The emissions are also represented as distributed variables. The Dynamics of the pollution stock is defined by a parabolic partial differential equation. We numerically characterize the feedback Nash equilibrium of a discrete-space model that still captures the spatial interactions among agents. We evaluate the impact of the strategic and spatially Dynamic behaviour of the agents on the design of equilibrium environmental policies.
Aline I. Maalouf - One of the best experts on this subject based on the ideXlab platform.
-
Finite Horizon H∞ Control for a Class of Linear Quantum Sampled-Data Measurement Systems: A Dynamic Game Approach*
IFAC Proceedings Volumes, 2011Co-Authors: Aline I. Maalouf, Ian R. PetersenAbstract:Abstract In this paper, the finite horizon H∞ control problem is solved for a class of linear quantum systems using a Dynamic Game approach for the case of sampled-data measurements. The methodology adopted involves a certain equivalence between the quantum problem and an auxiliary classical stochastic problem. Then, by solving the finite horizon H∞ control problem for the equivalent stochastic problem using some results from a corresponding deterministic problem following a Dynamic Game approach, the finite horizon H∞ control problem for the class of linear quantum systems under consideration is solved for the case of sampled-data measurements.
-
ACC - Finite horizon H ∞ control for a class of linear quantum measurement delayed systems: A Dynamic Game approach
Proceedings of the 2011 American Control Conference, 2011Co-Authors: Aline I. Maalouf, Ian R. PetersenAbstract:In this paper, a finite horizon H∞ control problem is solved for a class of linear quantum systems using a Dynamic Game approach for the case of delayed measurements. The methodology adopted involves an equivalence between the quantum problem and an auxiliary classical stochastic problem. Then, the finite horizon H∞ control problem for the class of linear quantum systems under consideration is solved for the case of delayed measurements by solving the finite horizon H∞ control problem for an equivalent stochastic H∞ control problem using some results from a corresponding deterministic problem following a Dynamic Game approach.
-
Finite horizon H ∞ control for a class of linear quantum systems: A Dynamic Game approach
Proceedings of the 2010 American Control Conference, 2010Co-Authors: Aline I. Maalouf, Ian R. PetersenAbstract:In this paper, the finite horizon H∞ control problem is solved for a class of linear quantum systems using a Dynamic Game approach. The methodology adopted involves an equivalence between the quantum problem and an auxiliary classical stochastic problem. Then, by solving the finite horizon H∞ control problem for the equivalent stochastic problem using results from a corresponding deterministic problem following a Dynamic Game approach, the finite horizon H∞ control problem for the class of linear quantum systems under consideration is solved.
