We propose a multiagent distributed actor-critic algorithm for multitask reinforcement learning (MRL), named Diff-DAC. The agents are connected, forming a (possibly sparse) network. Each agent is assigned a task and has access to data from this local task only. During the learning process, the agents are able to communicate some parameters to their neighbors. Since the agents incorporate their ...

Although reinforcement learning (RL) has been successfully deployed in a variety of tasks, learning speed remains a fundamental problem for applying RL in complex environments. Transfer learning aims to ameliorate this shortcoming by speeding up learning through the adaptation of previously learned behaviors in similar tasks. Transfer techniques often use an inter-task mapping, which determines...

This paper examines the development of a genetic adaptive fuzzy control system for the Inverted Pendulum. The inverted pendulum is a classical problem in Control Engineering, used for testing different control algorithms. The goal is to balance the inverted pendulum in the upright position by controlling the horizontal force applied to its cart. Because it is unstable and has a complicated nonl...

We consider the safety verification of controllers obtained via machine learning. This is an important problem as the employed machine learning techniques work well in practice, but cannot guarantee safety of the produced controller, which is typically represented as an artificial neural network. Nevertheless, such methods are used in safety-critical environments. In this paper we take a typica...

In this paper, we propose a novel adaptive dynamic programming (ADP) architecture with three networks, an action network, a critic network, and a reference network, to develop internal goalrepresentation for online learning and optimization. Unlike the traditional ADP design normally with an action network and a critic network, our approach integrates the third network, a reference network, int...

A switching control strategy for inverted pendulum systems is proposed based on the energy function and its transitions. We deal with a simplified second order model of cart-pendulum systems. Assume a virtual energy whose potential energy has an opposite sign. Paying attention to the motions of the pendulum, the energy changes are analyzed theoretically. According to these analyses, the conditi...

This paper obtains feedback stabilization of an inverted pendulum on a rotor arm by the “method of controlled Lagrangians”. This approach involves modifying the Lagrangian for the uncontrolled system so that the Euler-Lagrange equations derived from the modified or “controlled” Lagrangian describe the closed-loop system. For the closed-loop equations to be consistent with available control inpu...

Human movement is a “natural skill” employed to solve difficult problems in dynamics concerning the manipulation of a complex biomechanical system, the body, in an ever-changing environment. Continuous Interactive Simulation (CIS) is a technique that attempts to use this human capacity to solve problems in movement dynamics to solve problems concerning arbitrary dynamical systems. In this paper...