Abstract:Abstract-We present a general approach for online learning and optimal control of manipulation tasks in a supervisory teleoperation context, targeted to underwater remotely operated vehicles (ROVs). We use an online Bayesian nonparametric learning algorithm to build models of manipulation motions as task-parametrized hidden semi-Markov models (TP-HSMM) that capture the spatiotemporal characteristics of demonstrated motions in a probabilistic representation. Motions are then executed autonomously using an optim… Show more
“…Therefore, the generalization of the demonstrated trajectories to new situations is not concerned. Recent follow-up work [Havoutis and Calinon, 2017] addressed the online learning and the adaptation of the skill to new contexts by combining an optimal control approach and TP-GMM in [Calinon, 2015]. used ProMPs for incremental imitation with generalization to different contexts.…”
As robots and other intelligent agents move from simple environments and problems to more complex, unstructured settings, manually programming their behavior has become increasingly challenging and expensive. Often, it is easier for a teacher to demonstrate a desired behavior rather than attempt to manually engineer it. This process of learning from demonstrations, and the study of algorithms to do so, is called imitation learning. This work provides an introduction to imitation learning. It covers the underlying assumptions, approaches, and how they relate; the rich set of algorithms developed to tackle the problem; and advice on effective tools and implementation.We intend this paper to serve two audiences. First, we want to familiarize machine learning experts with the challenges of imitation learning, particularly those arising in robotics, and the interesting theoretical and practical distinctions between it and more familiar frameworks like statistical supervised learning theory and reinforcement learning. Second, we want to give roboticists and experts in applied artificial intelligence a broader appreciation for the frameworks and tools available for imitation learning.We organize our work by dividing imitation learning into directly replicating desired behavior (sometimes called behavioral cloning [Bain and Sammut, 1996]) and learning the hidden objectives of the desired behavior from demonstrations (called inverse optimal control [Kalman, 1964] or inverse reinforcement learning [Russell, 1998]). In addition to method analysis, we discuss the design decisions a practitioner must make when selecting an imitation learning approach. Moreover, application examples-such as robots that play table tennis [Kober and Peters, 2009] and programs that play the game of Go [Silver et al., 2016]-illustrate the properties and motivations behind different forms of imitation learning. We conclude by presenting a set of open questions and point towards possible future research directions.
“…Therefore, the generalization of the demonstrated trajectories to new situations is not concerned. Recent follow-up work [Havoutis and Calinon, 2017] addressed the online learning and the adaptation of the skill to new contexts by combining an optimal control approach and TP-GMM in [Calinon, 2015]. used ProMPs for incremental imitation with generalization to different contexts.…”
As robots and other intelligent agents move from simple environments and problems to more complex, unstructured settings, manually programming their behavior has become increasingly challenging and expensive. Often, it is easier for a teacher to demonstrate a desired behavior rather than attempt to manually engineer it. This process of learning from demonstrations, and the study of algorithms to do so, is called imitation learning. This work provides an introduction to imitation learning. It covers the underlying assumptions, approaches, and how they relate; the rich set of algorithms developed to tackle the problem; and advice on effective tools and implementation.We intend this paper to serve two audiences. First, we want to familiarize machine learning experts with the challenges of imitation learning, particularly those arising in robotics, and the interesting theoretical and practical distinctions between it and more familiar frameworks like statistical supervised learning theory and reinforcement learning. Second, we want to give roboticists and experts in applied artificial intelligence a broader appreciation for the frameworks and tools available for imitation learning.We organize our work by dividing imitation learning into directly replicating desired behavior (sometimes called behavioral cloning [Bain and Sammut, 1996]) and learning the hidden objectives of the desired behavior from demonstrations (called inverse optimal control [Kalman, 1964] or inverse reinforcement learning [Russell, 1998]). In addition to method analysis, we discuss the design decisions a practitioner must make when selecting an imitation learning approach. Moreover, application examples-such as robots that play table tennis [Kober and Peters, 2009] and programs that play the game of Go [Silver et al., 2016]-illustrate the properties and motivations behind different forms of imitation learning. We conclude by presenting a set of open questions and point towards possible future research directions.
“…The red dotted line is the motion predicted by the model from the current robot state. The gray square in the robot space represents what the robot end-effector state would be with a direct teleoperation behavior, which would only poorly match the current situation in the robot space (Color figure online) et al (2016) and Havoutis and Calinon (2017), we showed how such a model can be learned in an online manner, and be used in a teleoperation scenario with failing communication, to semi-autonomously perform an ROV task (hot-stabbing) using an MPC formulation for motion generation.…”
Teleoperation in domains such as deep-sea or space often requires the completion of a set of recurrent tasks. We present a framework that uses a probabilistic approach to learn from demonstration models of manipulation tasks. We show how such a framework can be used in an underwater ROV teleoperation context to assist the operator. The learned representation can be used to resolve inconsistencies between the operator's and the robot's space in a structured manner, and as a fall-back system to perform previously learned tasks autonomously when teleoperation is not possible. We evaluate our framework with a realistic ROV task on a teleoperation mock-up with a group of volunteers, showing a significant decrease in time to complete the task when our approach is used. In addition, we illustrate how the system can execute previously learned tasks autonomously when the communication with the operator is lost.
“…In (Farooq et al, 2016), the authors presented a state convergence-based control method with a Takagi-Sugeno (TS) fuzzy model for nonlinear teleoperation system. Havoutis et al (Havoutis et al, 2017) developed an integrated method involving optimal control and online learning to accomplish a manipulation task for underwater remotely operated vehicles during supervisory teleoperation. Additionally, Daniel et al proposed a user-controlled variable impedance method with implicit haptic feedback for unstructured environments (Walker et al , 2010).…”
To develop secure, natural, and effective teleoperation, the perception of the slave plays a key role for the interaction of a human operator with the environment. By sensing slave information, the human operator can choose the correct operation in a process of human-robot interaction. This paper develops an integrated scheme based on a hybrid control and virtual fixture approach for the telerobot. The human operator can sense the slave interaction condition and adjust the master device via the surface electromyographic signal. This hybrid control method integrates proportional-derivative control and variable stiffness control, and involves muscle activation at the same time. It is proposed to quantitatively analyse the human operator's control demand to enhance the control performance of the teleoperation system. In addition, due to unskilful operation and muscle physiological tremor of the human operator, a virtual fixture method is developed to ensure accuracy of operation and to reduce the operation pressure on the human operator. Experimental results demonstrated the effectiveness of the proposed method for the teleoperated robot.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.