RSS 2013 Workshop on Common Platforms in Robotic Manipulation

Description

This workshop will focus on common hardware and software platforms for robotic manipulation. Participants will present work related to this theme, including open source hardware, software and simulation platforms that are being made available to the community. We will discuss results from research projects demonstrating the successful (or unsuccessful) use of common platforms for manipulation, including open source, commercial, or government-furnished equipment. Topics of interest also include the notion of benchmarking and comparing available hardware and software platforms for grasping and manipulation through shared experimental protocols and data.

Our invited speakers will represent common platforms that have had a significant impact on robotic manipulation. In addition to including both hardware and software, we will aim to discuss the adoption and use of robotic platforms from both the point of view of the producer  and the user.

In addition to presentation and discussion sessions, the workshop will feature a Hands-On Session, with live demonstrations of novel designs intended for low-cost, easily manufacturable and widely available robotics hands.

Schedule

Full-day workshop, June 27th 2013 - in conjunction with RSS 2013, Berlin, Germany.

  • 08:45: Welcome
  • Session: Platforms I
  • 09:00 - Sachin Chitta (Willow Garage Inc.): "MoveIt!: A Common Platform for Motion Planning and Mobile Manipulation / PR2: Lessons learned from a Common Research Platform"
  • 09:30 - Martin Haegele (Fraunhofer IPA): "The Care-O-bot Family: Enable Real World Experiments with capable Service Robots"
  • 10:00 - Giorgio Metta (Istituto Italiano di Tecnologia): "iCub: a platform for research in embodied cognition"
  • 10:30 - Coffee Break
  • Session: Software
  • 11:00 - Domenico Prattichizzo (University of Siena): "SynGrasp: a flexible Matlab toolbox for human and robotic hands"
  • 11:30 - Stefan Schaal (Max-Planck-Insitute for Intelligent Systems and University of Southern California): "SL: A Simulation and Hard Real-Time Control Software Tool for Robotic Systems"
  • 12:00 - Lunch Break

Note: we have shifted the afternoon schedule to accomodate the lab tour and coffee break - we will start the afternoon sessions at 2pm and conclude the workshop at 5:30pm.

  • Hands-On Session with live hand demos
  • 02:00 - Lael Odhner (Yale University): The iRobot/Harvard/Yale ARM-H Hand, The OpenHand Model T
  • 02:15 - Matei Ciocarlie (Willow Garage, Inc.): The Velo Gripper
  • 02:30 - Raphael Deimel (TU Berlin): The RBU Hand
  • 02:45 - Matt Mason (Carnegie Mellon University): The MLab Hand
  • 03:00 - Live hand demonstrations featuring the four hand models presented above as well as the Kuka youBot
  • 03:30 - Coffee Break
  • Session: Platforms II
  • 04:00 - John Hsu (Open Source Robotics Foundation): "Gazebo simulation platform for robot software development,  lessons learned from DARPA Virtual Robotics Challenge."
  • 04:30 - Ana Murillo (University of Zaragoza): "A low-cost custom build robotic arm on the TurtleBot platform"
  • 04:45 - Marinus Danzer and Christian Scheurer (KUKA Laboratories GmbH): "youBot, iiwa, omniRob – KUKA reference platforms for robotics research and education"
  • 05:15 - Open Discussion
  • 05:30 - Wrap-up

Invited Speakers

Marinus Danzer and Christian Scheurer (KUKA Laboratories GmbH): "youBot, iiwa, omniRob – KUKA reference platforms for robotics research and education"


Sachin Chitta (Willow Garage): "MoveIt!: A Common Platform for Motion Planning and Mobile Manipulation / PR2: Lessons learned from a Common Research Platform"

In this talk, I will present MoveIt! - new software targeted at allowing you to build advanced applications integrating motion planning, kinematics, collision checking with grasping, manipulation, navigation, perception, and control. MoveIt! is robot agnostic software that can be quickly set up with your robot if a URDF representation of the robot is available. The MoveIt! Setup Assistant lets you configure MoveIt! for any robot, allowing you to visualize and interact with the robot model. MoveIt! can incorporate both actual sensor data and simulated models to build an environment representation. MoveIt! updates its representation of the environment on the fly, enabling reactive motion planning and execution, which is essential for applications in human-robot collaborative environments. Workspace analysis tools allow robot designers to test out the capabilities of their robot designs before building the hardware, using environment and object specific task specifications to quantify the workspace characteristics of different designs. 

I will also present lessons learned from the PR2 Beta Program, which has revolutionized research in mobile manipulation by making available a shared platform for software and hardware. I will point to some of the success stories where software tools developed as part of the program are now in widespread use throughout the community.


John Hsu (Open Source Robotics Foundation): "Gazebo simulation platform for robot software development,  lessons learned from DARPA Virtual Robotics Challenge."

Making robots perform useful tasks often requires coordination of various levels of modeling abstractions, from high level task planning to low level actuator control.  Maintaining consistently working hardware systems is a nontrivial and time consuming task, thus having a fast and efficient simulation platform that supports emulation of dynamics, sensor data generation, flexible software interface and visualization can help speed up design, implementation and testing processes.  While real-time robot simulations are still far from perfectly mimicking the real world behaviors of complex robotic systems, a well posed modeling tool can still aide researchers and software developers test algorithm implementations before it is dispatched on the real robot.


Giorgio Metta (Istituto Italiano di Tecnologia): "iCub: a platform for research in embodied cognition". 

I will present the iCub humanoid, a robotic platform designed for research in embodied cognition. At 104 cm tall, the iCub has the size of a three and half years old child. It can crawl on all fours and sit up to manipulate objects. Its hands have been designed to support sophisticate manipulation skills. The iCub is distributed as Open Source following the GPL/FDL licenses and can now count on a worldwide community of enthusiastic developers. The entire design is available for download from the project homepage and repository (http://www.iCub.org). About 25 robots have been built so far which are available in laboratories in Europe, US, and soon in Japan. It is one of the few platforms in the world with a sensitive full-body skin to deal with the physical interaction with the environment including possible people.


Ana Murillo (University of Zaragoza): "A low-cost custom build robotic arm on the TurtleBot platform"

This work presents a project that covers all the necessary steps to build a low-cost mobile platform equipped with a robotic arm. A Kinect-guided custom robotic arm on the TurtleBot platform provides the vacuum cleaner with basic manipulation capabilities. All stages in the project involve standard hardware components and open-source software modules. 1) The robotic arm is built with off-the-shelf parts and controlled from an Arduino board. 2) A Kinect sensor is used for object detection. 3) A ROS node manages all the elements, commanding the mobile base and designing targets to be grasped. The arm has been evaluated in order to identify the accuracy that can be expected with this setup, and a demonstration program, which simulates a simple cleaning task in which soft objects are moved out of the way, is described. Despite the simplicity of the setup and algorithms used, the experiments show notable robustness and performance from the system. This initial prototype has a simple pincer that enables grasping of easy objects only, while future work will address improvements for more advanced recognition and grasping strategies (full pdf).


Domenico Prattichizzo (University of Siena): "SynGrasp: a flexible Matlab toolbox for human and robotic hands"

SynGrasp is a MATLAB toolbox developed for the analysis of grasping, suitable both for robotic and human hands. It includes functions for the definition of hand kinematic structure and of the contact points with a grasped object. The coupling between joints induced by an underactuated control can be modeled. The hand modeling allows to define compliance at the contact, joint and actuator levels. The provided analysis functions can be used to investigate the main grasp properties: controllable forces and object displacement, manipulability analysis, grasp quality measures. Functions for the graphical representation of the hand, the object and the main analysis results are provided. 

The SynGrasp toolbox has been developed in the context of the EU Project “THE - The Hand Embodied” that aims to study how the embodied characteristics of the human hand and its sensors, the sensorimotor transformations, and the very constraints they impose, affect and determine the learning and control strategies we use for such fundamental cognitive functions as exploring, grasping and manipulating. For this reason, the toolbox provides several functions for human and robot grasping evaluation including specific functions for human hand synergies evaluation. 


Stefan Schaal (Max-Planck-Insitute for Intelligent Systems and University of Southern California): "SL: A Simulation and Hard Real-Time Control Software Tool for Robotic Systems"

Since about 20 years ago, born out of a real-time control multi-processor system using vxWorks, we have been creating the Simulation Lab (SL), a software tool that seamlessly realizes a physical simulator for robotics and a hard-real-time control system for physical robots, formerly using vxWorks, and now using Xenomai. SL has been used for many different robotics platforms, from Aldebaran NAOs, to Sensable Phantoms, Barrett WAM arms/hands, Sarcos Master/Save/Humanoids, KUKA LBR4, and various out platforms. Among other features, SL uses Featherstone algorithms for physical simulations, and also interfaces to ROS for non-real-time computations and interfaces. This talk will give an overview of the pros and cons of SL and anticipated future developments.

Contact

You can contact the organizers at coman13@willowgarage.com.

Organizers