social linkedin box blue 32
social facebook box blue 32
social twitter box blue 32
social facebook box blue 32

iit-advr-logo-v3

Events

Workshops

RO-MAN 2015

Workshop on Learning for Human-Robot Collaboration, 24th International Symposium on Robot and Human Interactive Communication (Ro-Man'2015)

ro man 2015Once isolated behind safety fences, the new emerging generation of robots endowed with more precise and sophisticated sensors, as well as better actuators, are materializing the idea of having robots working alongside people not only on manufacturing production lines, but also in spaces such as houses, museums, and hospitals. In this context, one of the next frontiers is the collaboration between humans and robots, which raises new challenges for robotics. A collaborative robot must be able to assist humans in a large diversity of tasks, understand its collaborator's intentions as well as communicate its own intentions, and decide when it can lead the task or when just follow its human counterpart. All these aspects demand the robot to be endowed with an adaptation capability so that it can satisfactorily collaborate with humans. In this sense, learning is a crucial feature for creating robots that can execute different tasks, and rapidly adapt to its human partner's actions and requirements. This workshop is aimed not only at bringing together researchers from human-robot interaction, robot control and machine learning fields, but also at creating valuable collaborations, and attracting more roboticists into this growing community.

Topics of interest include (but are NOT limited to) the following:

  • Learning from demonstration
  • Reinforcement learning
  • Active learning
  • Force and impedance control
  • Physical human-robot interaction
  • Human-robot coordination
  • Recognition and prediction of human actions
  • Reactive and proactive behaviors
  • Roles allocation
  • Haptic communication
  • Cooperative human-human interaction
  • Human activity understanding

Invited speakers:

Call for contributions:

        We welcome prospective participants to submit either full papers (up to 6 pages, to be orally presented) or extended abstracts (up to 2 pages, to be presented as poster). The manuscripts should use the IEEE RO-MAN two-column format. Please submit a PDF copy of your manuscript through our EasyChair platform at

https://easychair.org/conferences/?conf=lfhrc2015

      before May 22nd, 2015. Each paper will receive a minimum of two reviews. Papers will be selected based on their originality, relevance, contributions, technical clarity, and presentation. Accepted papers require that at least one of the authors register to the workshop.

Important dates:

Submission deadline: May 22, 2015
Notification of acceptance: July 15, 2015
Final submission: August 10, 2015
Workshop date: August 31, 2015

Accepted papers

  • "Performance Evaluation of Developed Intuitive Teaching Systems for Robot Manipulators", G. Capi, D. Hossain, S. Kaneko and K. Bode
  • "Learning Robot Skills from Human Feedback for Human-Robot Interaction Tasks", A. G. Kupcsik, D. Hsu and W. S. Lee
  • "Automated Interpretations of Unsupervised Learning-Derived Clusters for Activity Recognition", R. Freedman and S. Zilberstein
  • "Modeling of Human Postural Adaptations Using Reinforcement Learning", C. E. Sezener, L. Peternel, J. Camernik, J. Babic and E. Oztop
  • "Learning to Interact with Humans Using Goal-Directed and Habitual Behaviors", E. Renaudo, S. Devin, B. Girard, R. Chatila, R. Alami, M. Khamassi and A. Clodic

 

Tentative Program

Time Event
9:00 – 9:10 Opening by the organizers
9:10 – 9:45 Il Hong Suh
9:45 – 10:20 Heni Ben Amor
10:20 – 10:30 Poster teasers
10:30 – 10:50 Poster session + Coffee break
10:50 – 11:25 Dongheui Lee
11:25 – 12:00 Ayşe Küçükyılmaz
12:00 – 12:20 Discussion and wrap-up
12:20 – 14:00 Lunch break

Organizers:

leonel

Leonel Rozo (Main organizer)
Istituto Italiano di Tecnologia (IIT)
Via Morego, 30. 16163. Genoa, Italy.
Tel: +39 010 717 81 722
E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it
Web page: http://leonelrozo.weebly.com/

sylvain

Sylvain Calinon
Idiap Research Institute
Centre du Parc, Rue Marconi 19. CH-1920 Martigny, Switzerland.
Tel: +41 27 721 77 61
E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it
Web page: www.programming-by-demonstration.org/SylvainCalinon/

Last Updated on Monday, 20 July 2015 08:37

ICAR 2015 Workshop on Compliance Control in Legged Locomotion: Active vs. Passive

International Conference on Advanced Robotics (ICAR), Istanbul, Turkey

Workshop on Monday, 27th July 2015, Cibali Room, 9:00 - 17:30


Topic:

As noted in the seminal work of Zheng & Hemami (1985), a legged robotic system needs to exhibit compliant locomotion behavior while interacting with the environment, regardless of the given terrain type. Many experimental studies substantiated this hypothesis; the incorporation of compliance in legged locomotion is observed to be a key factor in managing the position/force trade-off to address environmental adaptation, disturbance rejection, and dynamically equilibrated gait generation.

With this in mind, this workshop aims to bring together researchers who developed active and/or passive compliance control schemes in various forms to improve legged locomotion behavior. A special emphasis will be given to the comparison between active and passive compliance control schemes. Questions, such as, “Up to what extent can software-controlled active compliance controllers emulate their physical counterparts” and “What are the unique advantages of passive compliance and it comes at what cost” will be chiefly discussed.

Motivation and Objectives:

Contemporary robotics technology allowed the use of highly improved hardware and control architectures, which led to exciting improvements in compliance robotics. For instance, various mechanical design topologies are successfully designed and implemented to address adjustable physical stiffness control, in an attempt to emulate certain behaviors of biological muscles for agile locomotion. That being said, some research groups adopted highly effective torque control schemes that enabled real-time adjustment of stiffness and damping without any physical elasticity. At this point, we have several solutions towards the same goal, thus, the time is right to discuss performance metrics, trade-offs and implementation difficulties of these solutions.

This workshop aims to bring together practitioners to discuss the key aspects of compliance control in legged locomotion. Our main motivation is to create a forum between researchers, so they can convey their own experiences to the community with presentation and discussions. A particular emphasis is given to the real-life implementation issues; the role of active/passive compliance will be discussed in different legged locomotion applications: humanoids, quadrupeds, and exoskeletons.

The following questions will be discussed.

  • It is argued that active compliance schemes can adjust the stiffness and damping properties, thus, enabling highly compliant robot behaviors. Up to what extent can they replace their physical counterparts?
  • Can active compliance controllers handle impulsive disturbances? How to handle delays caused by controllers, sensors etc? Long-term effects?
  • It is argued that passively compliant systems can emulate biological muscles in terms of efficient locomotion generation. Recent studies indicate that they cannot store/release high amounts of energy, therefore, is it true that they can lead to energy-efficient locomotion?
  • At what cost can passive compliance be implemented? What are the design and implementation difficulties? What are recent design trends?
  • In generating locomotion with adjustable impedance (active or passive), how to generate suitable impedance trajectories? What should be the cost function to minimize/maximize in defining target impedance characteristics?.

 
Schedule (scroll down for the talk abstracts)

9:00 Welcome
9:10 Barkan Ugurlu, Ozyegin University, Istanbul, Turkey
9:50 Claudio Semini, IIT Genova, Italy
10:30 coffee break
10:50 Nikos Tsagarakis, IIT Genova, Italy
11:30 Alexander Herzog, MPI Tuebingen, Germany
12:10 Marco Hutter, ETH Zurich, Switzerland
12:50 lunch break
14:30 Ramazan Unal, Vrije University, Belgium
15:10 Thiago Boaventura, ETH Zurich, Switzerland
15:50  coffee break
16:10 Erhan Oztop, Ozyegin University, Istanbul, Turkey
16:50 discussion
17:20 final remarks
17:30 end of workshop
 

Invited Speakers

NikosTsagarakis Nikos Tsagarakis, Humanoids and Human-Centered Mechatronics (Head), Dept. of Advanced Robotics, Istituto Italiano di Tecnologia (IIT), Italy

Talk: Efficient and Robust Compliant Actuation Arrangements for WALK-MAN humanoid
Abstract: The mechatronic development of humanoid systems has considerably progressed during the past three decades with various designs based on different actuation technologies, from motorized based systems to hydraulic and soft actuation technologies. Despite the advancements in the design of motorized humanoids/bipeds significant barriers remain, preventing robot hardware (physical structure and actuation) from equalling the performance of human in locomotion and full body motion in terms of physical robustness and efficiency. The talk will introduce ongoing work and results towards the development of our most recent humanoid platform WALK-MAN. Details on the actuation design approaches to achieve the necessary performance in terms of physical resilience, energy efficiency and high power and fast motion capabilities will be presented.

 

RamazanUnal Ramazan Unal, Dept. of Mechanical Engineering, Vrije University, Brussels, Belgium

Talk: Compliant actuators for lower limb exoskeletons and prostheses
Abstract: An increasing number of robotics researchers have realized that in animals and humans not only the brain creates the intelligence of the body, but that the morphology and biomechanics have a great impact on the way animals and humans think and move. A critical role in this respect is played by the neuro-mechanics of muscles, which have functional performance and control capabilities far in excess reached by artificial actuators. In traditional robotics, stiff actuators controlled as servomotor governed by the principle of “the stiffer the better”, produce high bandwidth control adapted to tasks requiring tracking of a desired trajectory with high accuracy. Many -mostly novel- applications requiring interaction with an unknown and dynamic environment including humans require dynamics that are not well suited to servomotors. Therefore, compliant actuators are being developed, which are inspired by biological motor control, where the influence of muscle spring-like properties and their control are of great importance. In the presentation we will highlight the role of compliant elements to design improved lower limb exoskeletons for assistance and rehabilitation and prostheses.


ErhanOztop Erhan Öztop, Dept. of Mechanical Engineering, Ozyegin University, Istanbul, Turkey

Talk: Central Nervous System Actively Modulates Muscle Impedance: Can we transfer skills from Humans to Impedance Controlled Robots?
Abstract: The sensorimotor learning capacity allows humans to efficiently learn to use new tools and control tasks. To realize this ability in complex systems such as robots is still elusive. Never the less, it is possible to use this human ability to transfer nontrivial sensorimotor skills to robots using a human-in-the-control-loop setup. The idea is to consider the target robotic platform as a tool that can be controlled by a human. Provided with an intuitive interface for controlling the robot, the human learns to perform a given task using the robot. After sufficient learning, the skilled control of the robot by the human provides learning data points that can be used to obtain an autonomous controller so that the robot can perform the task without human guidance. The feasibility of this framework is supported by the neuroscientific findings on body schema and has been shown to work for several robot skill synthesis scenarios. From an engineering point of view, the approach relies on techniques from teleoperation and machine learning, and has the same goals with robot learning by demonstration. The key difference is that the proposed framework includes the human in the control loop and employs the human brain as the adaptive controller to accomplish a given task. Once the control proficiency has been attained, the data generated by the human performance allows the human policy to be transferred to the robot. In this talk, I will introduce the human-in-the-loop framework and outline the current challenges for facilitating a wide impact for the development of adaptive systems for complex environments.

 

AlexanderHerzog Alexander Herzog, Autonomous Motion Department, Max-Planck-Institute for Intelligent Systems, Tuebingen, Germany

Talk: Active Compliance with LQR-based Momentum Control and Hierarchical Inverse Dynamics
Abstract: Hierarchical inverse dynamics based on cascades of quadratic programs have been proposed for the compliant control of legged robots. They have important benefits but to the best of our knowledge have never been implemented on a torque controlled humanoid where model inaccuracies, sensor noise and real-time computation requirements can be problematic. Using a reformulation of existing algorithms, we propose a simplification of the problem that allows to achieve real-time control. Momentum-based control is integrated in the task hierarchy and a LQR design approach is used to compute the desired associated closed-loop behavior and improve performance. Extensive experiments on various balancing and tracking tasks show very robust performance in the face of unknown disturbances, even when the humanoid is standing on one foot. Our results demonstrate that hierarchical inverse dynamics together with momentum control can be efficiently used for feedback control under real robot conditions. Although, easy to design in our balance experiments, desired momentum trajectories are non-trivial to chose when we tackle more sophisticated tasks. In the second part of my talk I will discuss preliminary simulation experiments that address this problem with a trajectory optimization approach.


MarcoHutter Marco Hutter, Inst. f. Robotik u. Intell. Syst., ETH, Zurich, Switzerland

Talk: Compliant actuation to enhance dynamic locomotion performance of legged robots
Abstract: Including mechanical compliance in the actuation system of legged robotic systems is known to provide several benefits such as precise torque controllability, inherent robustness, increase of peak power, and temporary storage of energy.  However, this comes at a price of more complex system dynamics and, due to the mechanical lowpass behavior, a decreasing control bandwidth – a problem that often scares people away from using highly compliant actuators in versatile robots. In this talk, I want to provide an insight into our experience with the application of highly compliant SEAs at the example of the (autonomous) quadrupedal robot StarlETH. The goal is to foster the understanding how to choose the compliance, how to exploit it using motion optimization and learning tools, and how to control it to achieve versatile behaviors.


ThiagoBoaventura Thiago Boaventura, Agile and Dexterous Robotics Lab., ETH, Zurich, Switzerland

Talk: Interaction force control in wearable robots
Abstract: To improve human quality of life is the main goal in the field of wearable robotics, which includes devices such as exoskeletons and limb-prostheses. These wearable robots are closely attached to human body in a symbiotic way, moving and working in tandem with the human. In several applications, it is paramount to control the robot forces applied to the user. In this talk I will give an overview of the commonly used approaches for interaction force control in wearable robots, and then propose a novel method which is currently being patented by ETH Zurich. This new control methods aims at improving the intrinsic limitations of traditional impedance control for masking inertia. Such approach is particularly suitable to provide an enhanced transparent (i.e. imperceptible) behavior for wearable robots.

 

BarkanUgurlu Barkan Ugurlu, Legged Locomotion Group (Head), Dept. of Mechanical Engineering, Ozyegin University, Istanbul, Turkey

Talk: Variable Ankle Stiffness Outperforms Optimal Constant Stiffness: Experiments on a Soft Exoskeleton
Abstract: Variable Stiffness Actuators (VSAs) have been around for quite some time; however, this crucial property has not been exploited in legged robot control task. In this talk, we present a real-time balance controller which utilizes the variable physical ankle stiffness by means of ZMP feedback. Inspired by the findings in biomechanics literature, a soft exoskeleton with pneumatic muscles were built and used as the testbed. The proposed variable ankle stiffness strategy was compared to optimal constant stiffness strategy. As the result, the proposed method outperformed the optimal constant stiffness method in terms of disturbance rejection, ZMP/CoM position regulation, and mechanical power; while air mass consumption rate increased about 70%. To the authors' knowledge, this is the first implementation of variable physical stiffness in the legged robot control literature. Details regarding this study will be discussed during the talk.

ClaudioSemini Claudio Semini, Dynamic Legged Systems (Head), Dept. of Advanced Robotics, Istituto Italiano di Tecnologia (IIT), Italy

Talk: Active impedance for increased Performance and Versatility of Legged Robots
Abstract: This talk will focus on the benefits of active impedance for versatile legged robots. To improve the performance and versatility of legged robots in real-world settings, it is crucial to properly control the interaction forces between the robot and the environment. We will focus on active impedance/compliance. Recent experiments on IIT’s fully torque-controlled hydraulic robot HyQ are used to illustrate the importance of active impedance for legged robots. HyQ has recently demonstrated a large set of behaviors ranging from highly dynamic motions to planned navigation on rough terrain, e.g. flying trot, punch bag disturbance rejection, and walking over stepping stones.


Organizers:


This workshop is organized and supported by the H2020 Topic Group on Field/Service Robots in unstructured Environments.
Futhermore this workshop is co-organized by the IEEE-RAS Technical Committee on Mechanisms and Design.

 

Last Updated on Saturday, 25 July 2015 20:46

IROS14 WS & IJHR SI on Whole-Body Control for Robots in the Real World

CALL FOR CONTRIBUTIONS to the twin events:



• Full-day Workshop at the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2014)

Organizers: F.L. Moro (IIT), M. Gienger (Honda RI-EU), O. Khatib (Stanford), E. Yoshida (AIST)



• Special Issue on the International Journal of Humanoid Robotics (IJHR) - June 2015

Editors: F.L. Moro (IIT), M. Gienger (Honda RI-EU), A. Goswami (Honda RI-US), O. Khatib (Stanford), E. Yoshida (AIST)



on the theme: “Whole-Body Control for Robots in the Real World”

Last Updated on Tuesday, 01 April 2014 10:41

Read more...

ICRA 2014 Workshop on Hydraulic Robots

Hydraulic Robots with Torque Control for Inverse
Dynamics and Active Compliance

2014 IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, China
This full-day workshop takes place on Sunday, June 1st, 2014. (SuWT7, room S229)

3rd workshop in the series of Hydraulic Robots (ICRA 2013, IROS 2013)

Organizers:

  • Claudio Semini, Istituto Italiano di Tecnologia (IIT)
  • Michele Focchi, Istituto Italiano di Tecnologia (IIT)
  • Thiago Boaventura, Swiss Federal Institute of Technology (ETH)
  • Jonas Buchli, Swiss Federal Institute of Technology (ETH)

Call for extended abstracts closed.

Final Schedule (scroll down for the talk abstracts)

9:00 - 9:30 Introduction and talk by Claudio Semini (Istituto Italiano di Tecnologia, Italy)
9:30 - 10:00 Talk by Jonas Buchli (ETHZ Zurich, Switzerland)
10:00 - 10:30 Talk by Thiago Boaventura (ETHZ Zurich, Switzerland)
10:30 - 10:50 coffee break
10:50 - 11:20 Talk by Michele Focchi (Istituto Italiano di Tecnologia, Italy)
11:20 - 11:50
Talk by Scott Kuindersma (MIT, USA)
11:50 - 12:30 Poster/Interactive session with teasers in the first 15min
12:30 - 14:00 lunch break
14:00 - 14:30 Talk by Jouni Mattila (Tampere University of Technology, Finland)
14:30 - 15:00
Talk by Hyungpil Moon (SungKyunKwan University, South Korea)
15:00 - 15:30 Discussion round
15:30 - 15:50 coffee break
15:50 - 16:20 Talk by
Ian Whiting (MOOG Controls Ltd., UK)
16:20 - 16:30 Concluding remarks

 

Abstract:

This full-day workshop will bring together key researchers of academia and industry to discuss the implementation and limitations of high-performance torque control, inverse dynamics and active compliance control of robots with hydraulic actuation. Inverse dynamics control and active compliance are prime examples to illustrate the benefits of torque control. Inverse dynamics control uses rigid-body dynamics models to calculate the joint torques required to achieve desired contact forces or motion. Active compliance (or better: impedance) is the implementation of virtual springs-dampers in either joint space or between any two points on the robot structure (e.g. hip and foot). This impedance can be adjusted in real-time according to the desired interaction behavior. For these reasons, torque control is critical for legged locomotion and manipulation in unstructured environments where a proper control of contact forces is crucial. Torque control is therefore a key technology for versatile service and field robots. Several companies, mostly funded through US defense grants (e.g. Boston Dynamics and Sarcos Raytheon), are developing robots and exoskeletons with impressive performance. All of them have torque control and hydraulic actuation and impressively demonstrate the advantages of this technology combination. Recently, also academic groups developed robots based on this technology (e.g. the quadruped robot HyQ). Our workshop is part of an academic and civil use oriented program focusing on the research, dissemination and application of this critical technology.
 

List of topics:

- Torque control on hydraulic robots
- Leg/arm impedance control
- Limitations on achievable impedance parameters in real implementations
- Different torque measurement approaches (strain gages, pressure difference, SEA)
- Applications of active compliance
- Inverse dynamics model-based control
- Feedback linearisation for hydraulic actuators
- Balancing of hydraulic legged robots
- Highly dynamic motions of hydraulic robots
- Whole body control of hydraulic robots
- Torque control in rehabilitation devices

List of Speakers

ClaudioSemini Claudio Semini, Istituto Italiano di Tecnologia (IIT), Italy, confirmed

Talk: Torque Control for increased Performance and Versatility of Legged Robots
Abstract: This talk will focus on the benefits of torque control for versatile legged robots. To improve the performance and versatility of legged robots in real-world settings, it is crucial to properly control the interaction forces between the robot and the environment. We will focus on active impedance/compliance. Recent experiments on IIT’s fully torque-controlled hydraulic robot HyQ are used to illustrate the importance of torque control for legged robots. HyQ has recently demonstrated a large set of behaviors ranging from highly dynamic motions to planned navigation on rough terrain, e.g. flying trot, punch bag disturbance rejection, and walking over stepping stones.


MicheleFocchi Michele Focchi, Istituto Italiano di Tecnologia (IIT), Italy, confirmed

Talk: High-slope Terrain Locomotion for Torque-Controlled Quadruped Robots
Abstract: Research into legged robotics is primarily motivated by the prospects of building machines that are able to navigate in challenging and complex environments that are predominantly non-flat. In this context, control of contact forces is fundamental to ensure stable contacts and stability of the robot. In my talk I will propose a planning/control framework for quasi-static walking for quadrupedal robots, implemented for a demanding application (walking inside 50deg V-shaped walls) in which regulation of ground reaction forces is crucial for the success of the walking behaviour. To achieve this the robot is distributing its weight among the stance legs so as to optimize user-defined criteria (e.g. no foot slippage and minimization of the actuators effort).

JouniMattila Jouni Mattila, Tampere University of Technology, Finland, confirmed

Talk: High Performance and Energy-efficient Torque Based Control of Hydraulic Manipulators
Abstract: This talk gives an overview of our recent developments in high performance torque based control of hydraulic manipulators. Our industrially driven objective is to develop high performance motion and interaction control systems for hydraulic manipulators that are energy-efficient. We present ways to improve four quadrant force-velocity operating envelope of hydraulic servodrives utilizing joint torque control, and how to couple variable displacement pumps with electric load sensing into overall energy-efficiency enhanced design. Our developments are founded on advanced nonlinear modeling and parameter adaptation coupled together with control strategy called Virtual Decomposition Control (VDC), which rigorously guarantees the stability of entire complex system without imposing additional approximations.

IanWhiting Ian Whiting, Moog Controls Ltd, Tewkesbury, UK, confirmed

Talk: Analysis by Dynamic Modelling – The Key to Understanding Miniature Hydraulic Servo Performance
Abstract: Whilst today’s industrial servo valves benefit from on-board sensors and digital electronics; the original Moog nozzle/flapper/feedback wire design remains the technology of choice where miniaturisation or extreme environments are required. This is evident in the recently developed and constantly evolving E024 series servo valve used for motorsport and robotic applications. Affordable computer based analysis tools today play a key role. This is especially true in servo actuation where dynamic analysis has become the norm both for product design and application. This presentation outlines the experiences of Moog in our everyday pursuance of design excellence.

ThiagoBoaventuraThiago Boaventura, Swiss Federal Institute of Technology (ETHZ), Switzerland, confirmed

Talk: Model-based design and analysis of hydraulic compliant controllers
Abstract: In order to obtain a high-performance actively compliant robot, it is desirable to design an inner torque closed-loop controller with as high bandwidth as possible. In this talk, I'll describe some intrinsic physical characteristics that limit the performance in force/torque control, and how to use model-based controllers to compensate for them with hydraulic actuators. Furthermore, I'll show that the torque control loop performance has a big impact in the overall leg stability when interacting with the environment, and also that other mechanical issues such as inertia and friction can strongly influence the robot compliance performance.

JonasBuchliJonas Buchli, Swiss Federal Institute of Technology (ETHZ), Switzerland, confirmed

Talk: It is the law...
Abstract: I will review the motivation behind the classical impedance control scheme with an inner torque/force control loop an an outer position control loop. I will show why this control scheme easily succeeds in providing overall robust robot performance in contact tasks (e.g. locomotion and manipulation) where other approaches fail. Even though this is a long established and well worked argument, surprisingly to this day a large fraction of the robotics community is ignoring this fundamental physics lesson. I will attempt to dispel the myth that good torque control is necessarily complicated and difficult to implement. And, what hardware it takes to do so. Finally, I will sketch out the tremendous opportunities that active impedance control offers when combined with model based and learning control. By understanding that this combination is key to versatile and robust robot performance, we will understand why (active) impedance control has been a bit of a forgotten treasure to this date.

ScottKuindersmaScott Kuindersma, MIT CSAIL, Cambridge, USA, confirmed

Talk: Achieving Stable Dynamic Locomotion with an Atlas Humanoid Robot
Abstract: What does it take to make a 155kg, 188cm tall hydraulic humanoid move dynamically? In this talk, I will describe our approach to achieving dynamic balancing, walking, and compliant upper body control with Atlas. Starting with an efficient optimization-based controller for whole-body locomotion, I will share our process of moving from simulation to the real robot, including system identification, state estimation, and joint-level control. Along the way l will highlight several properties of the system that make precise force control difficult and describe how we are coping with these limitations.

HyungpilMoonHyungpil Moon, SungKyunKwan University, Suwon, Korea

Talk: Limitations on force tracking control in hydraulic robot systems with servo-valves
Abstract: Robot control technologies have been developed mainly for electrical motor driven robot systems. As opposed to velocity control of robot joints, force control can performs better in situations where manipulators make contact with handling objects or environments. In case of robots with electric motors, we may directly measure joint torques from various joint torque sensors or directly measure the contact force from force/torque sensor at the tip of end effector. Current control is also applicable to realize force control for mechanically well-designed robotic systems. Because of dynamically fast responses of electric motors, many force control methodologies have been applied to robotic systems without too much attention to the electric responses of the motor itself. Although the electric-hydraulic analogy is generally accepted in many analysis problems of both, the force control of hydraulic robot system with servo-valve is not the case. In this workshop, we will discuss the issue of force tracking in a simply hydraulic system. The fundamental limitations of the hydraulic system with servo-valves require advanced control algorithms as a necessity tool. We will revisit the more-than-a-decade-old issue and remind ourselves of what limitations we are dealing with.

Poster Presentations

Feng Gao, Chenkun Qi, Qiao Sun, Xianbao Chen and Xinghua Tian:
 "Hydraulic system design for the quadruped robot 'Baby Elephant'"

Sven Hubacher, Thiago Boaventura, Farbod Farshidian, Michele Focchi, Claudio Semini, and Jonas Buchli:
 "Optimization of the Low-Level Torque Controller of the Quadruped Robot HyQ"

J. Meng, Y. Li, H. Zhang, H. Chai and G. Zhang:
 "Stair Climbing and Person Tracking of SCalf-II"

Alexander Winkler, Ioannis Havoutis and Claudio Semini:
 "Combining Active Compliance with on-the-fly Motion Replanning using Haptic Information"

P. Beeson, J. James, J. Kramer, F. Weng, K. Khokar, M. Chu, S. Gee, M. Rynderman, E. Huber, D. Kortenkamp and R. Burridge:
"Development on Atlas Humanoid Robot for Remote Execution of Disaster Relief Tasks under Human Supervisory Control"



Open Call for 1-page abstracts (closed)
 
We invite researchers that are working on the control and/or design of high-performance hydraulic robots, components and robotic devices to submit a 1-page abstract. Authors of accepted abstracts will be given the opportunity to present their work during the poster session, preceded by a short poster teaser presentation.

Important Dates

1-page abstracts submission deadline: April 25th, 2014 (23:59 PST)
Notification of acceptance: April 30th, 2014

Submission Instructions 

Please prepare a 1-page extended abstract PDF using the IEEE template (US-letter!).
The template can be downloaded here: http://www6.cityu.edu.hk/icra2014/paper_submission.htm
Before submitting check your pdf with the IEEE compliance test for ICRA 2014. Then submit your extended abstract to Michele Focchi (michele.focchi{a}iit.it ) with the tag [ICRA2014] in the subject line.
Abstracts will be checked by the organizers for suitability for this workshop and completeness, but will not be fully peer-reviewed.


Atlas
Motorsport applications of the Moog E024 Servo valve – High performance, precision miniaturization and extreme environmental robustnessHyQ Robotmoon data

Last Updated on Friday, 30 May 2014 06:51

INFORMATION NOTICE ON COOKIES

IIT's website uses the following types of cookies: browsing/session, analytics, functional and third party cookies. Users can choose whether or not to accept the use of cookies and access the website.
By clicking on further information, the full information notice on the types of cookies used will be displayed and you will be able to choose whether or not to accept cookies whilst browsing on the website.

Try our new site and tell us what you think
Take me there