scholarly journals Manipulation of free-floating objects using Faraday flows and deep reinforcement learning

2022 ◽  
Vol 12 (1) ◽  
Author(s):  
David Hardman ◽  
Thomas George Thuruthel ◽  
Fumiya Iida
Keyword(s):  
Reinforcement Learning ◽  
Control Problem ◽  
Large Scale ◽  
Open Loop ◽  
Dimensional Parameter ◽  
Fluid Medium ◽  
Surface Flows ◽  
Loop Control ◽  
Floating Object ◽  
Floating Objects

AbstractThe ability to remotely control a free-floating object through surface flows on a fluid medium can facilitate numerous applications. Current studies on this problem have been limited to uni-directional motion control due to the challenging nature of the control problem. Analytical modelling of the object dynamics is difficult due to the high-dimensionality and mixing of the surface flows while the control problem is hard due to the nonlinear slow dynamics of the fluid medium, underactuation, and chaotic regions. This study presents a methodology for manipulation of free-floating objects using large-scale physical experimentation and recent advances in deep reinforcement learning. We demonstrate our methodology through the open-loop control of a free-floating object in water using a robotic arm. Our learned control policy is relatively quick to obtain, highly data efficient, and easily scalable to a higher-dimensional parameter space and/or experimental scenarios. Our results show the potential of data-driven approaches for solving and analyzing highly complex nonlinear control problems.

Adjoint-Based Optimization Procedure for Active Vibration Control of Nonlinear Mechanical Systems

2017 ◽  
Vol 139 (8) ◽  
Author(s):  
Carmine M. Pappalardo ◽  
Domenico Guida
Keyword(s):  
Optimal Control ◽  
Control Theory ◽  
Control Problem ◽  
Numerical Solution ◽  
Optimal Control Theory ◽  
Numerical Procedure ◽  
Nonlinear Optimal Control ◽  
Open Loop ◽  
Adjoint Equations ◽  
Loop Control

In this paper, a new computational algorithm for the numerical solution of the adjoint equations for the nonlinear optimal control problem is introduced. To this end, the main features of the optimal control theory are briefly reviewed and effectively employed to derive the adjoint equations for the active control of a mechanical system forced by external excitations. A general nonlinear formulation of the cost functional is assumed, and a feedforward (open-loop) control scheme is considered in the analytical structure of the control architecture. By doing so, the adjoint equations resulting from the optimal control theory enter into the formulation of a nonlinear differential-algebraic two-point boundary value problem, which mathematically describes the solution of the motion control problem under consideration. For the numerical solution of the problem at hand, an adjoint-based control optimization computational procedure is developed in this work to effectively and efficiently compute a nonlinear optimal control policy. A numerical example is provided in the paper to show the principal analytical aspects of the adjoint method. In particular, the feasibility and the effectiveness of the proposed adjoint-based numerical procedure are demonstrated for the reduction of the mechanical vibrations of a nonlinear two degrees-of-freedom dynamical system.

scholarly journals Parametric Approach to Trajectory Tracking Control of Robot Manipulators

2013 ◽  
Vol 2013 ◽  
pp. 1-6
Author(s):  
Shijie Zhang ◽  
Yi Ning
Keyword(s):  
Optimal Control ◽  
Optimal Control Problem ◽  
Control Problem ◽  
Trajectory Tracking ◽  
Tracking Control ◽  
Optimal Trajectory ◽  
Open Loop ◽  
Parametric Approach ◽  
Trajectory Tracking Control ◽  
Loop Control

The mathematic description of the trajectory of robot manipulators with the optimal trajectory tracking problem is formulated as an optimal control problem, and a parametric approach is proposed for the optimal trajectory tracking control problem. The optimal control problem is first solved as an open loop optimal control problem by using a time scaling transform and the control parameterization method. Then, by virtue of the relationship between the optimal open loop control and the optimal closed loop control along the optimal trajectory, a practical method is presented to calculate an approximate optimal feedback gain matrix, without having to solve an optimal control problem involving the complex Riccati-like matrix differential equation coupled with the original system dynamics. Simulation results of 2-link robot manipulator are presented to show the effectiveness of the proposed method.

scholarly journals Pneumatic Servo Bearing Actuator with Multiple Bearing Pads for Ultraprecise Positioning

2013 ◽  
Vol 7 (5) ◽  
pp. 498-505 ◽  
Author(s):  
Shuhei Tsujimura ◽  
◽  
Yusuke Hashimoto ◽  
Takashi Matsuoka ◽  
Tomoko Hirayama ◽  
...  
Keyword(s):  
Research And Development ◽  
Dynamic Characteristics ◽  
Large Scale ◽  
Thrust Bearing ◽  
Open Loop ◽  
Open Loop Control ◽  
Positioning Accuracy ◽  
Loop Control ◽  
Positioning Systems ◽  
The Difference

With the increase in demand for semiconductor products, ultrafine linear patterning technologies for Large-Scale Integrations (LSIs) have been making progress. The requested positioning accuracy in such ultraprecise apparatuses is of nanometer order. To meet such specific needs, the research and development of a variety of actuators has been necessary. Our laboratory has developed a ‘Pneumatic Servo Bearing Actuator (PSBA),’ a novel actuator that uses pneumatic servo technology for ultraprecise positioning. Our past studies have showed that the minimum resolution of PSBA was almost 6 nm, even under open loop control; thus, we concluded that the PSBA was a promising actuator in advanced ultraprecise positioning systems. However, the stroke of the PSBA was comparatively short. To expand the stroke of the actuator, we proposed and developed a new PSBA with multiple thin thrust-bearing pads. The main purpose in this study is to investigate the positioning properties of the new PSBA with multiple thrust bearing pads. The obtained characteristics of the PSBA can be enumerated as follows. (1) The PSBA with multiple bearing pads achieves a longer stroke than the PSBA with a single pad. (2) The difference in the thickness of the bearing pads affects the dynamic characteristics of the actuator. (3) The minimum positioning resolution of the developed PSBA with twenty-nine bearing pads is about 2 – 4 nm with feedback control.

scholarly journals Existence and uniqueness of constrained globally optimal feedback controls in a linear-quadratic framework

2006 ◽  
Vol 2006 ◽  
pp. 1-22 ◽  
Author(s):  
Yashan Xu
Keyword(s):  
Optimal Control ◽  
Optimal Control Problem ◽  
Control Problem ◽  
Existence And Uniqueness ◽  
Closed Loop ◽  
Synthesis Method ◽  
Open Loop ◽  
Linear Quadratic ◽  
Feedback Controls ◽  
Loop Control

A constrained closed-loop optimal control problem is considered in a linear-quadratic framework. To solve the problem, a special type open-loop optimal control problem and a standard open-loop optimal control problem are introduced and carefully studied, via which the existence and uniqueness of the globally optimal closed-loop control is established by a synthesis method.

Decomposition Solution of Open-Loop 'Cheap' Control Problem

1993 ◽  
Author(s):  
S.X. Shen ◽  
V.G. Gourishankar ◽  
Q. Xia ◽  
M. Rao
Keyword(s):  
Control Problem ◽  
Open Loop ◽  
Cheap Control
Sign in / Sign up

Export Citation Format

Share Document