Zobrazeno 1 - 10
of 194
pro vyhledávání: '"A. Abdolshah"'
Autor:
Fortunić, Edmundo Pozo, Yildirim, Mehmet C., Ossadnik, Dennis, Swikir, Abdalla, Abdolshah, Saeed, Haddadin, Sami
Elastic actuation taps into elastic elements' energy storage for dynamic motions beyond rigid actuation. While Series Elastic Actuators (SEA) and Variable Stiffness Actuators (VSA) are highly sophisticated, they do not fully provide control over ener
Externí odkaz:
http://arxiv.org/abs/2309.07873
Autor:
Hamad, Mazin, Gutierrez-Moreno, Jesus, Kussaba, Hugo T. M., Mansfeld, Nico, Abdolshah, Saeed, Swikir, Abdalla, Burgard, Wolfram, Haddadin, Sami
This paper proposes a framework for generating fast, smooth and predictable braking manoeuvers for a controlled robot. The proposed framework integrates two approaches to obtain feasible modal limits for designing braking trajectories. The first appr
Externí odkaz:
http://arxiv.org/abs/2306.06525
Autor:
Gupta, Sunil, Shilton, Alistair, A V, Arun Kumar, Ryan, Shannon, Abdolshah, Majid, Le, Hung, Rana, Santu, Berk, Julian, Rashid, Mahad, Venkatesh, Svetha
In this paper we introduce BO-Muse, a new approach to human-AI teaming for the optimization of expensive black-box functions. Inspired by the intrinsic difficulty of extracting expert knowledge and distilling it back into AI models and by observation
Externí odkaz:
http://arxiv.org/abs/2303.01684
Autor:
Ossadnik, Dennis, Yildirim, Mehmet C., Wu, Fan, Swikir, Abdalla, Kussaba, Hugo T. M., Abdolshah, Saeed, Haddadin, Sami
Publikováno v:
2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 3536-3543
Compliance in actuation has been exploited to generate highly dynamic maneuvers such as throwing that take advantage of the potential energy stored in joint springs. However, the energy storage and release could not be well-timed yet. On the contrary
Externí odkaz:
http://arxiv.org/abs/2212.14741
Autor:
Le, Hung, George, Thommen Karimpanal, Abdolshah, Majid, Nguyen, Dung, Do, Kien, Gupta, Sunil, Venkatesh, Svetha
We introduce a constrained optimization method for policy gradient reinforcement learning, which uses a virtual trust region to regulate each policy update. In addition to using the proximity of one single old policy as the normal trust region, we pr
Externí odkaz:
http://arxiv.org/abs/2204.09315
The current technical specification ISO/TS15066:2016(E) for safe human-robot interaction contains logically conflicting definitions for the contact between human and robot. This may result in different interpretations for the contact classification a
Externí odkaz:
http://arxiv.org/abs/2203.02706
We introduce a novel training procedure for policy gradient methods wherein episodic memory is used to optimize the hyperparameters of reinforcement learning algorithms on-the-fly. Unlike other hyperparameter searches, we formulate hyperparameter sch
Externí odkaz:
http://arxiv.org/abs/2112.01853
Episodic control enables sample efficiency in reinforcement learning by recalling past experiences from an episodic memory. We propose a new model-based episodic memory of trajectories addressing current limitations of episodic control. Our memory es
Externí odkaz:
http://arxiv.org/abs/2111.02104
Autor:
Karimpanal, Thommen George, Le, Hung, Abdolshah, Majid, Rana, Santu, Gupta, Sunil, Tran, Truyen, Venkatesh, Svetha
The optimistic nature of the Q-learning target leads to an overestimation bias, which is an inherent problem associated with standard $Q-$learning. Such a bias fails to account for the possibility of low returns, particularly in risky scenarios. Howe
Externí odkaz:
http://arxiv.org/abs/2111.02787
Autor:
Kirschner, Robin Jeanne, Mayer, Henning, Burr, Lisa, Mansfeld, Nico, Abdolshah, Saeed, Haddadin, Sami
Publikováno v:
in IEEE Robotics and Automation Letters, vol. 7, no. 2, pp. 2993-3000, April 2022
In robotics, many control and planning schemes have been developed to ensure human physical safety in human-robot interaction. The human psychological state and the expectation towards the robot, however, are typically neglected. Even if the robot be
Externí odkaz:
http://arxiv.org/abs/2109.07201