| 1. |
|
|
| 2. |
|
|
| 3. |
|
|
| 4. |
|
|
| 5. |
- Montebelli, Alberto, et al.
(author)
-
On handing down our tools to robots : Single-phase kinesthetic teaching for dynamic in-contact tasks
- 2015
-
In: Proceedings of the 2015 IEEE International Conference on Robotics and Automation. - : IEEE conference proceedings. - 9781479969234 - 9781479969210 ; , s. 5628-5634
-
Conference paper (peer-reviewed)abstract
- We present a (generalizable) method aimed tosimultaneously transfer positional and force requirements en-coded in a physical human skill (wood planing) from a humaninstructor to a robotic arm through kinesthetic teaching. Weachieve our goal through a novel use of a common sensoryconfiguration, constituted by a force/torque sensor mountedbetween the tool and the flange of a robotic arm. The roboticarm is endowed with integrated torque sensors at each joint.The mathematical model used to capture the general dynamicof the interaction between the human user and the wood surfaceis based on Dynamic Movement Primitives. During reenactmentof the task, the system can imitate and generalize the demon-strated spatial requirements, as well as their associated forceprofiles. Therefore, the robotic arm acquires the capacity toreproduce the dynamic profile for in-contact tasks requiringan articulated coordination in the distribution of forces. Forexample, the capacity to effectively operate the plane on a woodplank over multiple strokes, according to the demonstration ofthe human instructor.
|
|
| 6. |
- Montebelli, Alberto, et al.
(author)
-
On Transferring Crafting Intentions from Humans to Robots : A Message to Manufacturers of the (Near) Future
- 2018
-
In: Advances in Manufacturing Technology XXXII. - Amsterdam : IOS Press. - 9781614999010 - 9781614999027 ; , s. 163-168
-
Conference paper (peer-reviewed)abstract
- Our present work aligns three results from previous robotics research in simultaneous kinesthetic teaching of spatial and force/torque requirements for “in-contact” tasks, to highlight the endeavor towards the creation of safe, flexible, cost effective, confidential, natural programming interfaces, a crucial tool for the manufacturing domain of the future. The tasks that we here consider overarch different dimensions of complexity, from writing with a marker on a white slate to using a wood plane. Eventually, incrementally assisted kinesthetic teaching (IAKT) allows human experts to refine their demonstrations under modulated robotic assistance, thus converging, by a limit process constituted of a sequence of sub-perfect individual demonstrations, towards the “ideal” crafting intention, i.e. the humanly unreachable, perfect execution of the task. In the closing discussion, we demonstrate how this approach can find space in contemporary industrial and SMSE manufacturing, in order to aim for improved production quality and performance.
|
|
| 7. |
- Montebelli, Alberto, et al.
(author)
-
Reframing HRI Education : A Dialogic Reformulation of HRI Education to Promote Diverse Thinking and Scientific Progress
- 2017
-
In: Journal of Human-Robot Interaction. - : Association for Computing Machinery (ACM). - 2163-0364. ; 6:2, s. 3-26
-
Journal article (peer-reviewed)abstract
- Over the last few years, technological developments in semi-autonomous machines have raised awareness about the strategic importance of human-robot interaction (HRI) and its technical and social implications. At the same time, HRI still lacks an established pedagogic tradition in the coordination of its intrinsically interdisciplinary nature. This scenario presents steep and urgent challenges for HRI education. Our contribution presents a normative interdisciplinary dialogic framework for HRI education, denoted InDia wheel, aimed toward seamless and coherent integration of the variety of disciplines that contribute to HRI. Our framework deemphasizes technical mastery, reducing it to a necessary yet not sufficient condition for HRI design, thus modifying the stereotypical narration of HRI-relevant disciplines and creating favorable conditions for a more diverse participation of students. Prospectively, we argue, the design of an educational 'space of interaction’ that focuses on a variety of voices, without giving supremacy to one over the other, will be key to successful HRI education and practice.
|
|
| 8. |
- Racca, Mattia, et al.
(author)
-
Learning in-contact control strategies from demonstration
- 2016
-
In: IROS 2016. - : IEEE. - 9781509037629 ; , s. 688-695
-
Conference paper (peer-reviewed)abstract
- Learning to perform tasks like pulling a door handle or pushing a button, inherently easy for a human, can be surprisingly difficult for a robot. A crucial problem in these kinds of in-contact tasks is the context specificity of pose and force requirements. In this paper, a robot learns in-contact tasks from human kinesthetic demonstrations. To address the need to balance between the position and force constraints, we propose a model based on the hidden semi-Markov model (HSMM) and Cartesian impedance control. The model captures uncertainty over time and space and allows the robot to smoothly satisfy a task's position and force constraints by online modulation of impedance controller stiffness according to the HSMM state belief. In experiments, a KUKA LWR 4+ robotic arm equipped with a force/torque sensor at the wrist successfully learns from human demonstrations how to pull a door handle and push a button.
|
|
| 9. |
- Steinmetz, Franz, et al.
(author)
-
Simultaneous kinesthetic teaching of positional and force requirements for sequential in-contact tasks
- 2015
-
In: Proceedings of the 2015 IEEE-RAS International Conference on Humanoid Robots (Humanoids). - : IEEE Computer Society. - 9781479968855 - 9781479968848 ; , s. 202-209
-
Conference paper (peer-reviewed)abstract
- This paper demonstrates a method for simulta-neous transfer of positional and force requirements for in-contact tasks from a human instructor to a robotic arm throughkinesthetic teaching. This is achieved by a specific use of thesensory configuration, where a force/torque sensor is mountedbetween the tool and the flange of a robotic arm endowedwith integrated torque sensors at each joint. The humandemonstration is modeled using Dynamic Movement Primitives.Following human demonstration, the robot arm is provided withthe capacity to perform sequential in-contact tasks, for examplewriting on a notepad a previously demonstrated sequence ofcharacters. During the reenactment of the task, the systemis not only able to imitate and generalize from demonstratedtrajectories, but also from their associated force profiles. In fact,the implemented framework is extended to successfully recoverfrom perturbations of the trajectory during reenactment andto cope with dynamic environments.
|
|
| 10. |
|
|
