This paper addresses the problem of online inverse reinforcement learning for systems with limited data and uncertain dynamics. In the developed approach, the state and control trajectories are recorded online by observing an agent perform a task, and reward function estimation is performed in real-time using a novel inverse reinforcement learning approach. Parameter estimation is performed concurrently to help compensate for uncertainties in the agent's dynamics. Data insufficiency is resolved by developing a data-driven update law to estimate the optimal feedback controller. The estimated controller can then be queried to artificially create additional data to drive reward function estimation.
Inverse Reinforcement Learning (RL) can be used to determine the behavior of Space Objects (SOs) by estimating the reward function that an SO is using for control. The approach discussed in this work can be used to analyze maneuvering of SOs from observational data. The inverse RL problem is solved using maximum causal entropy. This approach determines the optimal reward function that a SO is using while maneuvering with random disturbances by assuming that the observed trajectories are optimal with respect to the SO's own reward function. Lastly, this paper develops results for scenarios invo…
This paper addresses the problem of online inverse reinforcement learning for nonlinear systems with modeling uncertainties while in the presence of unknown disturbances. The developed approach observes state and input trajectories for an agent and identifies the unknown reward function online. Sub-optimality introduced in the observed trajectories by the unknown external disturbance is compensated for using a novel model-based inverse reinforcement learning approach. The observer estimates the external disturbances and uses the resulting estimates to learn the dynamic model of the demonstrato…
In this paper, we propose a fast reinforcement learning (RL) control algorithm that enables online control of large-scale networked dynamic systems. RL is an effective way of designing model-free linear quadratic regulator (LQR) controllers for linear time-invariant (LTI) networks with unknown state-space models. However, when the network size is large, conventional RL can result in unacceptably long learning times. The proposed approach is to construct a compressed state vector by projecting the measured state through a projective matrix. This matrix is constructed from online measurements of…
In this paper, we propose a model-free adaptive learning solution for a model-following control problem. This approach employs policy iteration, to find an optimal adaptive control solution. It utilizes a moving finite-horizon of model-following error measurements. In addition, the control strategy is designed by using a projection mechanism that employs Lagrange dynamics. It allows for real-time tuning of derived actor-critic structures to find the optimal model-following strategy and sustain optimized adaptation performance. Finally, the efficacy of the proposed framework is emphasized throu…
We consider the inverse problem of dynamic games, where cost function parameters are sought which explain observed behavior of interacting players. Maximum entropy inverse reinforcement learning is extended to the N-player case in order to solve inverse dynamic games with continuous-valued state and control spaces. We present methods for identification of cost function parameters from observed data which correspond to (i) a Pareto efficient solution, (ii) an open-loop Nash equilibrium or (iii) a feedback Nash equilibrium. Furthermore, we give results on the unbiasedness of the estimation of co…
We propose a new reinforcement learning based approach to designing hierarchical linear quadratic regulator (LQR) controllers for heterogeneous linear multi-agent systems with unknown state-space models and separated control objectives. The separation arises from grouping the agents into multiple non-overlapping groups, and defining the control goal as two distinct objectives. The first objective aims to minimize a group-wise block-decentralized LQR function that models group-level mission. The second objective, on the other hand, tries to minimize an LQR function between the average states (c…
Model-based Vol/VAR optimization method is widely used to eliminate voltage violations and reduce network losses. However, the parameters of active distribution networks(ADNs) are not onsite identified, so significant errors may be involved in the model and make the model-based method infeasible. To cope with this critical issue, we propose a novel two-stage deep reinforcement learning (DRL) method to improve the voltage profile by regulating inverter-based energy resources, which consists of offline stage and online stage. In the offline stage, a highly efficient adversarial reinforcement lea…
In this paper, we consider the problem of computing parameters of an objective function for a discrete-time optimal control problem from state and control trajectories with active control constraints. We propose a novel method of inverse optimal control that has a computationally efficient online form in which pairs of states and controls from given state and control trajectories are processed sequentially without being stored or processed in batches. We establish conditions guaranteeing the uniqueness of the objective-function parameters computed by our proposed method from trajectories with …
Inferring the intent of an intelligent agent from demonstrations and subsequently predicting its behavior, is a critical task in many collaborative settings. A common approach to solve this problem is the framework of inverse reinforcement learning (IRL), where the observed agent, e.g., a human demonstrator, is assumed to behave according to an intrinsic cost function that reflects its intent and informs its control actions. In this work, we reformulate the IRL inference problem to learning control Lyapunov functions (CLF) from demonstrations by exploiting the inverse optimality property, whic…
Skolkovo Institute of Science and Technology（斯科尔科沃科学技术研究院）
Online reinforcement learning is concerned with training an agent on-the-fly via dynamic interaction with the environment. Here, due to the specifics of the application, it is not generally possible to perform long pre-training, as it is commonly done in off-line, model-free approaches, which are akin to dynamic programming. Such applications may be found more frequently in industry, rather than in pure digital fields, such as cloud services, video games, database management, etc., where reinforcement learning has been demonstrating success. Online reinforcement learning, in contrast, is more …
The distributed Volt/Var control (VVC) methods have been widely studied for active distribution networks(ADNs), which is based on perfect model and real-time P2P communication. However, the model is always incomplete with significant parameter errors and such P2P communication system is hard to maintain. In this paper, we propose an online multi-agent reinforcement learning and decentralized control framework (OLDC) for VVC. In this framework, the VVC problem is formulated as a constrained Markov game and we propose a novel multi-agent constrained soft actor-critic (MACSAC) reinforcement learn…
Online ride-hailing services have become a prevalent transportation system across the world. In this paper, we study a challenging problem of how to direct vacant taxis around a city such that supplies and demands can be balanced in online ride-hailing services. We design a new reward scheme that considers multiple performance metrics of online ride-hailing services. We also propose a novel deep reinforcement learning method named Deep-Q-Network with Action Mask (AM-DQN) masking off unnecessary actions in various locations such that agents can learn much faster and more efficiently. We conduct…
University of Ottawa（渥太华大学）；Aswan University（阿斯旺大学）；University of Texas at Arlington（德克萨斯大学阿灵顿分校）
The control problem of the flexible wing aircraft is challenging due to the prevailing and high nonlinear deformations in the flexible wing system. This urged for new control mechanisms that are robust to the real-time variations in the wing's aerodynamics. An online control mechanism based on a value iteration reinforcement learning process is developed for flexible wing aerial structures. It employs a model-free control policy framework and a guaranteed convergent adaptive learning architecture to solve the system's Bellman optimality equation. A Riccati equation is derived and shown to be e…
