Lately, nonlinear modelpredictivecontrol (NMPC) has been successfully applied to (semi-) autonomous driving problems and has proven to be a very promising technique. However, accurate controlmodels for real vehicle...
详细信息
Lately, nonlinear modelpredictivecontrol (NMPC) has been successfully applied to (semi-) autonomous driving problems and has proven to be a very promising technique. However, accurate controlmodels for real vehicles could require costly and time-demanding specific measurements. To address this problem, the exploitation of system data to complement or derive the prediction model of the NMPC has been explored, employing learning dynamics approaches within learning-based NMPC (LbNMPC). Its application to the automotive field has focused on discrete gray-box modeling, in which a nominal dynamics model is enhanced by the data-driven component. In this manuscript, we present an LbNMPC controller for a real go-kart based on a continuous black-box model of the accelerations obtained by Gaussian processes (GP). We show the effectiveness of the proposed approach by testing the controller on a real go-kart vehicle, highlighting the approximation steps required to get an exploitable GP model on a real-time application.
This paper proposes a method to encourage safety in modelpredictivecontrol (MPC)-based Reinforcement learning (RL) via Gaussian Process (GP) regression. The framework consists of 1) a parametric MPC scheme that is e...
详细信息
ISBN:
(纸本)9781713872344
This paper proposes a method to encourage safety in modelpredictivecontrol (MPC)-based Reinforcement learning (RL) via Gaussian Process (GP) regression. The framework consists of 1) a parametric MPC scheme that is employed as model-basedcontroller with approximate knowledge on the real system's dynamics, 2) an episodic RL algorithm tasked with adjusting the MPC parametrization in order to increase its performance, and 3) GP regressors used to estimate, directly from data, constraints on the MPC parameters capable of predicting, up to some probability, whether the parametrization is likely to yield a safe or unsafe policy. These constraints are then enforced onto the RL updates in an effort to enhance the learning method with a probabilistic safety mechanism. Compared to other recent publications combining safe RL with MPC, our method does not require further assumptions on, e.g., the prediction model in order to retain computational tractability. We illustrate the results of our method in a numerical example on the control of a quadrotor drone in a safety-critical environment. Copyright (c) 2023 The Authors. This is an open access article under the CC BY-NC-ND license (https://***/licenses/by-nc-nd/4.0/)
This paper introduces a predictive congestion pricing method in cities wherein the tolls alter from region to region. We consider a large urban network is partitioned into multiple regions each with a well-defined Mac...
详细信息
This paper introduces a predictive congestion pricing method in cities wherein the tolls alter from region to region. We consider a large urban network is partitioned into multiple regions each with a well-defined Macroscopic Fundamental Diagram (MFD) where multiple routes exist between each origin and destination regions. The proposed cordon pricing method is designed to (i) minimize vehicles' total time spent in the network and (ii) aim for a revenue-neutral tolling. A controller based on modelpredictivecontrol (MPC) approach is proposed to determine the (possibly negative) optimal time-and region-varying tolls. The MPC controller comprises a regional MFD-based traffic model with no need of destination information and a long-short term memory neural network (LSTM-NN) to obtain an accurate estimation of inter-region transfer flows. Results of numerical experiments indicate the effectiveness of the proposed congestion pricing method to achieve the two objectives simultaneously, compared with No toll and reactive feedback controllers.
This paper proposes a method to encourage safety in modelpredictivecontrol (MPC)-based Reinforcement learning (RL) via Gaussian Process (GP) regression. The framework consists of 1) a parametric MPC scheme that is e...
详细信息
This paper proposes a method to encourage safety in modelpredictivecontrol (MPC)-based Reinforcement learning (RL) via Gaussian Process (GP) regression. The framework consists of 1) a parametric MPC scheme that is employed as model-basedcontroller with approximate knowledge on the real system's dynamics, 2) an episodic RL algorithm tasked with adjusting the MPC parametrization in order to increase its performance, and 3) GP regressors used to estimate, directly from data, constraints on the MPC parameters capable of predicting, up to some probability, whether the parametrization is likely to yield a safe or unsafe policy. These constraints are then enforced onto the RL updates in an effort to enhance the learning method with a probabilistic safety mechanism. Compared to other recent publications combining safe RL with MPC, our method does not require further assumptions on, e.g., the prediction model in order to retain computational tractability. We illustrate the results of our method in a numerical example on the control of a quadrotor drone in a safety-critical environment.
暂无评论