Abstract:Several approaches have been proposed to assist humans in co-manipulation and teleoperation tasks given demonstrated trajectories. However, these approaches are not applicable when the demonstrations are suboptimal or when the generalization capabilities of the learned models cannot cope with the changes in the environment. Nevertheless, in real co-manipulation and teleoperation tasks, the original demonstrations will often be suboptimal and a learning system must be able to cope with new situations. This pape… Show more
“…Ewerton et al [12] tackle the problem of dynamic environment states, e.g. moving obstacle, by learning the parameters of a Gaussian process that outputs a ProMP distribution in realtime based on the current state of the environment.…”
Section: Related Workmentioning
confidence: 99%
“…where the first two terms are equal to the proposed controller in (12). We assume that the eigenvalues of the matrix Σ s , which is derived from the cross-correlation between two consecutive time steps, converge to zero as the duration of the time step tends towards zero.…”
Section: B Single Modementioning
confidence: 99%
“…In [7], demonstrated skills are modeled as the result of linear second-order dynamics. These approaches have later been extended to multimodality [8]- [11] and the realtime adaptation to context variables [12], [13].…”
In the context of learning from demonstration (LfD), trajectory policy representations such as probabilistic movement primitives (ProMPs) allow for rich modeling of demonstrated skills. To reproduce a learned skill with a real robot, a feedback controller is required to cope with perturbations and to react to dynamic changes in the environment. In this paper, we propose a generalized probabilistic control approach that merges the probabilistic modeling of the demonstrated movements and the feedback control action for reproducing the demonstrated behavior. We show that our controller can be easily employed, outperforming both original controller and a controller with constant feedback gains. Furthermore, we show that the proposed approach is able to solve dynamically changing tasks by modeling the demonstrated behavior as Gaussian mixtures and by introducing context variables. We demonstrate the capability of the approach with experiments in simulation and by teaching a 7-axis Franka Emika Panda robot to drop a ball into a moving box with only few demonstrations.
“…Ewerton et al [12] tackle the problem of dynamic environment states, e.g. moving obstacle, by learning the parameters of a Gaussian process that outputs a ProMP distribution in realtime based on the current state of the environment.…”
Section: Related Workmentioning
confidence: 99%
“…where the first two terms are equal to the proposed controller in (12). We assume that the eigenvalues of the matrix Σ s , which is derived from the cross-correlation between two consecutive time steps, converge to zero as the duration of the time step tends towards zero.…”
Section: B Single Modementioning
confidence: 99%
“…In [7], demonstrated skills are modeled as the result of linear second-order dynamics. These approaches have later been extended to multimodality [8]- [11] and the realtime adaptation to context variables [12], [13].…”
In the context of learning from demonstration (LfD), trajectory policy representations such as probabilistic movement primitives (ProMPs) allow for rich modeling of demonstrated skills. To reproduce a learned skill with a real robot, a feedback controller is required to cope with perturbations and to react to dynamic changes in the environment. In this paper, we propose a generalized probabilistic control approach that merges the probabilistic modeling of the demonstrated movements and the feedback control action for reproducing the demonstrated behavior. We show that our controller can be easily employed, outperforming both original controller and a controller with constant feedback gains. Furthermore, we show that the proposed approach is able to solve dynamically changing tasks by modeling the demonstrated behavior as Gaussian mixtures and by introducing context variables. We demonstrate the capability of the approach with experiments in simulation and by teaching a 7-axis Franka Emika Panda robot to drop a ball into a moving box with only few demonstrations.
“…TbD is an efficient approach to reduce the complexity of teaching a robot to perform new tasks (Billard et al, 2008 ; Yang et al, 2018 ). With this approach, a human tutor demonstrates how to implement a task to a robot easily (Ewerton et al, 2019 ). Then, the robot learns the key features from human demonstration and repeats it by itself.…”
Though a robot can reproduce the demonstration trajectory from a human demonstrator by teleoperation, there is a certain error between the reproduced trajectory and the desired trajectory. To minimize this error, we propose a multimodal incremental learning framework based on a teleoperation strategy that can enable the robot to reproduce the demonstration task accurately. The multimodal demonstration data are collected from two different kinds of sensors in the demonstration phase. Then, the Kalman filter (KF) and dynamic time warping (DTW) algorithms are used to preprocessing the data for the multiple sensor signals. The KF algorithm is mainly used to fuse sensor data of different modalities, and the DTW algorithm is used to align the data in the same timeline. The preprocessed demonstration data are further trained and learned by the incremental learning network and sent to a Baxter robot for reproducing the task demonstrated by the human. Comparative experiments have been performed to verify the effectiveness of the proposed framework.
Cardeñoso Fernandez, Franklin; Caarls, Wouter (Advisor). Deep reinforcement learning for haptic shared control in unknown tasks. Rio de Janeiro, 2020. 123p.
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.