Force-position control for robot manipulators has become a basic control regime for advanced applications such as interaction or cooperative tasks that imply contact to environment or external object. Thus, for many a...
详细信息
Actor-Critic (AC) architecture has the salient feature, for the plethora of Reinforcement Learning schemes, that two intertwining neural networks (NN) collaborate to deploy a motor learning mechanism that oversees and...
详细信息
A multi-fingered robotic hand with curved fingertips enables contact re-positioning without reattaching at the expense of fingertip rolling. This rolling stands for a characteristic that facilitates dexterous manipula...
详细信息
A multi-fingered robotic hand with curved fingertips enables contact re-positioning without reattaching at the expense of fingertip rolling. This rolling stands for a characteristic that facilitates dexterous manipulation but results in an algebraically complex dynamic model subject to such constraints. The hemispherical shape of fingertips allows a dexterous manipulation when controlling the tangent forces, which are essential to rotate object. However, the measurement of the object angle in practice requires tactile-optical sensing. In this paper, considering robotic fingers with curved soft tips, we propose a feedback control that ensures optimal dynamical grasping of a circular rigid object. It is shown that the collaboration of the contact forces, to get a minimum pose of internal forces, and the tangential forces, to induce the conditions for assuring the grasp closure, is necessary to get a skillful manipulation. In this case, the orientation control of a circular object to the desired angle while avoiding direct measurement of the object angle is presented. Stability conditions of the system are presented in the sense of stability-in-the-manifold. Finally, representative simulations are shown and discussed.
Actor-Critic (AC) architecture has the salient feature, for the plethora of Reinforcement Learning schemes, that two intertwining neural networks (NN) collaborate to deploy a motor learning mechanism that oversees and...
Actor-Critic (AC) architecture has the salient feature, for the plethora of Reinforcement Learning schemes, that two intertwining neural networks (NN) collaborate to deploy a motor learning mechanism that oversees and evaluates the action to control a dynamical system as the robot manipulator. Since such NNs can be studied as dynamical systems to learn how to control the robot dynamics with a given performance, it has finally paved the way to deal with stability analysis, unfortunately, few works have addressed it. In this paper, we propose a Critic-NN whose approximation of the value function substantiates the decision making mechanism that collaborates to tune the Actor-NN action (approximation of inverse dynamics). The novel proposed design of the adaptation of the neural weights yields Lyapunov stability that provides explicit conditions for an attractive invariant set that render a stable regime using a quite simple NN with one hidden layer. Numerical simulations show the performance of the proposed approach. In addition, robustness is analyzed when the robot is subject to Liptchitz disturbances, interestingly showing relaunching of the learning mechanism when needed. Finally, a discussion on dealing with asymptotic stability, robustness issues, and the learning mechanism from a reward provided by the expert user is addressed.
Force-position control for robot manipulators has become a basic control regime for advanced applications such as interaction or cooperative tasks that imply contact to environment or external object. Thus, for many a...
详细信息
ISBN:
(纸本)9781665494090
Force-position control for robot manipulators has become a basic control regime for advanced applications such as interaction or cooperative tasks that imply contact to environment or external object. Thus, for many applications environmental information is evaluated continuously to generate adequate control actions to complete the task. There has been proposed techniques such as adaptive or intelligent control to cover some of these requirements, however it seems insufficient given the uncertainty involved when dealing with the environment. The emergence of deep learning has recently renewed interest in these tasks from the perspective of reinforcement learning algorithms. In this paper, an actor-critic scheme is proposed for simultaneous control of contact force and joint positions while moving along an environmental rigid surface. The critic stage evaluates information of the environment through an extended error manifold, trying to minimize the cumulative future rewards by a neural network. The actor stage, which receives environment information from the critic stage, is used to approximate inverse robot dynamics. Together with two continuous orthogonal sliding PID controllers gives rise to simultaneous tracking of position and force, with stability analysis. Numerical simulations show the performance of the proposed approach under several conditions.
The actor-critic scheme stands for a powerful algorithm to design controllers for linear and non-linear systems subject to changing or highly uncertain dynamics. In particular, the actor-critic scheme that has succeed...
详细信息
ISBN:
(纸本)9781665408745
The actor-critic scheme stands for a powerful algorithm to design controllers for linear and non-linear systems subject to changing or highly uncertain dynamics. In particular, the actor-critic scheme that has succeeded is typically based on two neural network stages in a hierarchical architecture where the critic stage approximates the reward cost function. In contrast, the dynamic of the system is estimated by another neural network in the actor stage. This paper proposes an adaptive actor-critic robot learning on a lower dimension invariant error manifold as part of the Performance Evaluator. The proposed scheme guarantees an envelope of exponential convergence of tracking errors using a modified Lyapunov function, throughout integral sliding mode enforced for all time, where this becomes fundamental to drive also the learning of Reward function. Simulations show a non-linear dynamical robot learning tracking a time-varying trajectory under this Reinforcement Learning scheme.
This study was performed to support an expeditionary team of Colombian scientists seeking to promote and undertake research in various scientific fields in *** work was part of the Colombian Antarctic program,which co...
详细信息
This study was performed to support an expeditionary team of Colombian scientists seeking to promote and undertake research in various scientific fields in *** work was part of the Colombian Antarctic program,which comprises various projects intended to lead to a permanent Colombian scientific base being established in the *** first step involved installing a Colombian-made wind turbine to produce electricity at a permanent *** aeolian turbine was designed to provide sufficient electricity to illuminate and heat a small *** turbine was constructed using readily accessible materials but taking the Madrid Protocol environmental regulations into *** project was performed at the Argentinian Antarctic Marambio Station on Seymour Island,off the Antarctic *** initial field study performed in 2015 was the first of three phases of the *** the initial phase,local meteorological data were gathered to support development of a prototype turbine and to allow a design to be selected that was robust enough for the extreme environmental *** wind turbine was then constructed in *** second phase involved transporting the turbine to Antarctica and installing it at the Marambio Station in ***,the physical conditions of structural and electronic components of the turbine were carefully inspected at the beginning of 2020(after the turbine had operated continually for^2 years)to allow repairs to be made and any necessary re-engineering to be performed.
作者：
Schulte, DPSkolnick, AHe has supported the development and operation of several naval systems
including advanced component selection for Trident II fire control and navigation systems. He served as branch manager of the Surface Ship ASW Combat System Branch which acted as the acquisition engineering agent for the AN/SQQ-89 Surface Ship Anti-Submarine Warfare Weapon System. He was then selected to manage the Module Engineering Department which provided engineering support to numerous naval systems including the AN/BSY-1 Submarine Combat System and the Trident II fire control and navigation system. He then served as the deputy program manager for NAVSEA Progressive Maintenance (2M/ATE). He holds a B.S. degree in Electrical Engineering from Purdue University and currently is pursuing a Maste's degree in Public Environmental Affairs at Indiana University—Purdue University
Indianapolis. He served at Applied Physics Laboratory/The Johns Hopkins University in missile development
then aboard USS Boston (CAG-1) and played leading roles in several weapon system developments (Regulus Terrier Tartar Talos) inertial navigation (Polaris) deep submergence (DSRV) and advanced ship designs (SES). He later was director Combat System Integration Naval Sea Systems Command and head Combat Projects Naval Ship Engineering Center. He led the Navy's High Energy Lasers and Directed Energy Weapons development efforts. He was vice president advanced technology at Operations Research Inc. and vice president maritime engineering at Defense Group Inc. before starting SSC in 1991. Dr. Skolnick holds a B.S. degree in Mathematics and Economics
Queens College an M.A. degree in Mathematics and Philosophy Columbia University an M.S. degree in Electrical/Aeronautical Engineering U.S. Naval Postgraduate School and a Ph.D. in Electrical Engineering and Applied Mathematics from Polytechnic University in New York. He is the author of many published papers on engineering design issues source selection procedures and large-scale complex technology problems
The Fleet continues to require high performance systems that can operate with dependability in the seas' unforgiving environments and under hostile action. Those demands are not new. What has changed is the urgent...
详细信息
The Fleet continues to require high performance systems that can operate with dependability in the seas' unforgiving environments and under hostile action. Those demands are not new. What has changed is the urgent priority formerly assigned to national defense issues. The arguments for continued superpower military strength are now roiled in politics along with unsettled budgets and uncertain force level projections. Current expectations revolve about indefinite fiscal and operational issues (difficult funding constraints and broadband threats). In the actual event of ''doing more with less,'' a practical response is to apply the creative power available from sound engineering judgement and the crucible of experience to the immediate needs of the Fleet. The attempt to shorten the path between advanced development effort and Fleet use has been tried occasionally in the past, often, without exemplary results. The Sustainable Hardware and Affordable Readiness Practices (SHARP) program, is a generic R&D effort under OpNav sponsorship that has been working steadily on sensible solutions to product engineering problems. Armed today with fast-time, large-scale computation abilities and modern tools for technical problem solving coupled with specialized engineering knowledge, it has been refreshed and is underway satisfying existing Fleet needs. The relationship between fully responsive engineering services and current operational needs is always demanding. The connection between advanced engineering development (6.3 category funds) and immediate Fleet usage brings added complexity and challenge, both technical and organizational. Illustrative examples of affordable engineering solutions to ''retain, revise, replace or retire'' questions are presented within the context of both Fleet realities and budgetary limitations. The discussion covers legacy system support, civil/military considerations and Fleet maintenance issues. It describes the substantial and critical payoffs i
作者：
KATZ, RSJAHNKE, LJEWETT, CECdr. Larry Jahnke
USN:is presently Head of the Architecture Branch of the Avionics Engineering division AIR-546 of the Naval Air Systems Command. Among his current responsibilities is to lead implementation activities of the NAVAIR Advanced Avionics Architecture study described in this paper. Cdr. Jahnke graduated from the University of Minnesota with a B.S. degree in aeronautical engineering and was commissioned in 1974. After flight training as a Naval flight officer he was assigned to Naval Air Station Barbers Point Hawaii where he served as Tactical Coordinator for P-3B aircraft. He was assigned to the Communications Directorate of the Joint Staff in 1990 where he participated in support of Desert Shield/Desert Storm and was part of the original cadre of officers responsible for the “C41 for the Warrior” concept. Cdr. Jahnke also has a Master of Science degree from the University of Southern California and is a 1990 graduate of the Industrial College of the Armed Forces.Cdr. Charles E. Jewett
USN:is currently the Common Avionics Requirements Officer for Naval Aircraft Programs. He has served the Navy as an Aeronautical Engineering Duty Officer since 1982 with previous defense acquisition assignments as the Avionics Architecture and Engineering Branch Head Fighter/Attack Avionics systems Engineering Branch Head and A-12 Avionics Officer and A-6F Deputy Program Manager and the A-6 Avionics Officer. Cdr. Jewett entered the Navy as an Aviation Officer Candidate in 1971 receiving his commission and earning his wings as a Naval Flight Officer the same year. After graduating from the U.S. Naval Test Pilot School in 1976 he was assigned to the Strike Aircraft Test Directorate of the Naval Air Test Center where he participated in various electronic warfare electro-optics and software update evaluations for A-6 EA-6B and OV-10 aircraft. In Cdr. Jewett's previous assignment at NAVAIRSYSCOM he led a major Avionics Architecture Study (the subject of this paper) that surveyed cutting-edge avionics technol
To establish a planning basis for future avionics systems, the Naval Air Systems Command (NAVAIR) conducted an avionics architecture investigation during 1992-1993, culminating in a final report published in August 19...
To establish a planning basis for future avionics systems, the Naval Air Systems Command (NAVAIR) conducted an avionics architecture investigation during 1992-1993, culminating in a final report published in August 1993. In the course of the study, U.S. Industry provided significant information to a NAVAIR avionics database for both technologies and systems integration methods. From the study emerged an implementation strategy to allow NAVAIR to develop effective avionics systems in the future that use commercial products and standards where applicable but also allow the ready use of new and emerging technologies. Recommended strategies concentrate on the development process, especially the use of sound systems engineering techniques and the maximum practical use of commercial standards and products. This paper reviews the methodology employed during the NAVAIR investigation, and presents the key findings and resulting implementation strategies. The paper concludes with a brief summary of current implementation plans at NAVAIR.
The military services are being moved in the direction of performance-based specifications and standards. They are being steered against dictating ''how to'' produce an item since such action foreclose...
The military services are being moved in the direction of performance-based specifications and standards. They are being steered against dictating ''how to'' produce an item since such action forecloses on the ability to gain access to components or technology that may have a commercial equivalent. Why should the engineering community embrace the new approach? Aside from the obvious weight of it being approved policy, therefore currently mandated, it warrants examination because it is the correct approach at this time when applied to appropriate products. Military specifications and standards are to be displaced then, by acceptable alternative contractor design solutions. Industry bidders will be allowed to propose the particular design details, permitting procurement flexibility by contractually citing only system level or interface requirements, both physical and functional. Hopefully, this can broaden the industrial base and increase competition with reduced costs to follow. Conceptually, the approach appears both performance-sensible and cost-attractive (there are, of course, consequent risks) but how does implementation proceed? Is it possible to pursue the goals envisioned along paths that are not in themselves experimental? Can the American postulate, minimal loss of life and limb to U.S. military people, continue to be honored? Experience and track record elsewhere imply encouraging possibilities in select situations-useful prospects are identified and discussed in practical terms.
暂无评论