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


Events ■ ICRA 2013 WS on Simulation Tools

ICRA 2013 Workshop on Developments of Simulation Tools for Robotics & Biomechanics

2013 IEEE International Conference on Robotics and Automation (ICRA), Karlsruhe, Germany.

The workshop is scheduled to take place on Friday, May 10, 2013.


  • Houman Dallali (Department of Advanced Robotics, Istituto Italiano di Tecnologia, Italy)
  • Gustavo Medrano-Cerda (Department of Advanced Robotics, Istituto Italiano di Tecnologia, Italy)
  • Paul Fisette (The Centre for Mechatronics (CEREM) at the Université catholique de Louvain, Belgium)
  • Nikolaos Tsagarakis (Department of Advanced Robotics, Istituto Italiano di Tecnologia, Italy)
  • Jochen Steil (The Research Institute for Cognition and Robotics (COR-Lab) at The Universität Bielefeld, Germany)
  • Darwin G. Caldwell (Department of Advanced Robotics, Istituto Italiano di Tecnologia, Italy)

This workshop aims at bringing together the researchers in robotics and related areas to share and discuss the novel tools and technologies in the area of simulation software. As the robotic systems are becoming more complex and the performance requirements on such systems are getting tighter, simulation studies become a crucial part of developing such complex systems. For instance emerging fields in robotics such as robots with variable stiffness and variable impedance actuators need new features in simulation tools that were not necessary when dealing with classical stiff or rigid robots. Hence, the main purpose of this workshop is to review and discuss the existing simulation tools and to exchange views on the required future development of these tools in order to cater for the new generation of biologically inspired robots with variable stiffness and impedance actuators.


COMAN Simulator - IIT   

CoMan Snapshot1

DartsLab - JPL

DARTSLAB Pathfinder


Gazeob1Gazeob - OSRF

Oncilla simulator

 V-REP simulator



Simulation tools play a significant role in design and development of complex robotics systems. Emergent areas in the field of robotics such as design and development of compliant robots, or variable impedance robots need dynamic simulation tools, to model the complex behavior of such robots and to help in understanding the use of compliance and elasticity for superior motion performance in humans and animals.

Moreover, as the performance of the robots is being pushed further simple control methods such as PID is not sufficient and more advanced "model-based multi-variable control methods (especially for robots with elastic or variable impedance actuators) are required to fully exploit their motion capabilities and to obtain the best possible performance. In order to implement such control algorithms a dynamic simulator which models the actuator dynamics of the next generation of robots is essential.
In addition, as new robots are being developed or upgraded the dynamic simulation tool should be able to keep up with these rapid changes and provide a user friendly interface to quickly model the changes in the robot, and automatically generate the code for dynamic equations of the robot. The method of code generation itself (" e.g. Newton-Euler or Lagrange-Newton methods) versus learned models (e.g. with Gaussian Process Regression") has an impact on the number of equations representing the robot dynamics and hence on the speed of the simulation. All these features are desirable in the next generation of simulation software, which are often lacking form the existing tools.

Modeling and simulation of rigid robots has been relatively matured over the past two decades, however in order to model the state of the art in robotics with new actuators and advanced control algorithms as well as realistic physics there is a need to have improved simulation tools as evident by the Gazebo project to develop a tool for DARPA [1].

On the other hand, open source models and sharing the code is one of the most important solutions to increase the interaction among robotic research groups and to provide a common tool for testing various algorithms on a single environment. OpenHRP [2], the recent Gazebo and ROS projects are two of the latest projects following this approach.


stryk    pic radkhah bunt
1. Oskar von Stryk & Katayon Radkhah, TU Darmstadt, Germany;
Title: Dynamic Modeling for Simulation, Optimization & Control of Bio-Inspired Elastic Robots.
For design and control of elastically actuated robots accounting for their highly nonlinear dynamic motion behavior is essential. In this talk we will discuss the role of and methodologies applied for the development of the bio-inspired, tendon-driven elastic robot arm BioRob and the humanoid jogging robot BioBiped. Furthermore, the  requirements for a coherent and possibly open source framework for dynamic modeling, optimization (e.g. parameter calibration and trajectory optimization) and model-based control and possible steps towards its development are discussed.

2. Nathan Koenig, Open Source Robotics Foundation, Co-founder of Gazebo, USA,
Title: The many faces of simulation: Use cases for a general purpose simulator.
Simulation tools exist for a wide variety of domains and use cases, including legged locomotion, underwater autonomous vehicles, soft body simulation, and aerial vehicles. These use cases generally rely on separate and specialized simulation engines capable of generating the correct physical and visual properties required to test and develop hardware and software for their respective domain. As multi-purpose robots interact with each other and people, the need for a scalable simulator with a wide range of capabilities increases. As an open source and community supported platform, Gazebo is ready to meet these needs. The flexible architecture used by Gazebo supports rapid prototyping, regression testing, research studies involving human robot interaction, and person simulation. The near future will see the integration of legged robots, large scale outdoor environment, and cloud based simulation. While specialized simulators will always be required, a general purpose simulator will benefit a large sector of the research and development community and open the door to new opportunities and collaboration.

3. Shin'ichiro Nakaoka, The National Institute of Advanced Industrial Science and Technology (AIST), Japan,
Title: Choreonoid: Extensible GUI Framework for Handling Virtual Robots.
Choreonoid is a software framework for implementing graphical robotics applications. Visualizations and graphical user interfaces (GUIs) for handling virtual robot models can be efficiently implemented as plugins running on the framework. The framework has been released as an open source software so that everyone can use it with its source code. Dynamics simulation of robots and their controllers can be performed on Choreonoid. In addition to that, various functions such as choreographing robot motions and doing motion planning have been implemented and such various functions can be flexibly integrated to work together. The Choreonoid framework is well-designed to achieve that.

Abhi Jain
4. Abhi Jain, Jet Propulsion Laboratory, DARTS Lab., USA,
Title: Structure Based Modeling and Computational Architecture for Robotic Systems.
We describe a computational architecture for meeting a diverse range of robot modeling needs encompassing analysis, simulation and embedded modeling for robotic systems. The architecture builds upon the spatial operator algebra theoretical framework for computational dynamics. It allows applications to meet the broad range of computational modeling needs coherently and with fast, structure-based computational algorithms. The paper describes the SOA computational architecture, the DARTS computational dynamics software, and application modeling layers.

5. Marc Freese, Coppelia Robotics, Developer of V-REP, Switzerland,
Title: Distributed Control in Robotics Simulations: a Case Study.
Distributed control with respect to robotics simulation is a recurrent theme in robotics and plays an important role. There is, however, little discussion regarding the various distributed control techniques and their efficient implementation. As the problem is inherently complex, there is an inherent trade-off to most approaches, and offering several approaches in a robot simulator gives the user a maximum of freedom and flexibility, especially if those approaches can work hand-in-hand. This work illustrates various examples around the V-REP robot simulator: based from ground up on a distributed control architecture, it robustly fits complex scenarios in which a diversity of distributed control techniques operate side-by-side. This allows for versatile applications, including rapid algorithm development, systems verification, safety/remote monitoring, training and education, and factory automation simulation.

6. Paul Fisette, CEREM, UCL, Belgium,
Title: ROBOTRAN, Self-contained Multibody Models for Robot Modeling and Real-Time Control.
Development of a realistic and precise mathematical model of robots of any type (manipulator, walking, rolling, etc.) plays a key role in the design and testing of controllers, and in trajectory generators, before experimentation on the real system. Moreover, the growing requirements with respect to dynamic performances currently make the design of robots more challenging and more complex. Whereas geometrical specifications and basic kinematic capabilities can be reasonably obtained via standard computer tools and via dedicated experimentation on prototypes, 3D kinematics and dynamics require more advanced mathematical formulations which are generally impossible to obtain analytically. In this respect, the multibody approach is perfectly suited to produce those models, the purpose of this discipline being to generate the kinematic and dynamic models of any articulated system, like mechanisms, robots, vehicles or the human body. The features of the ROBOTRAN program, developed at UCL, are described in the present paper. This software is a symbolic generator of multibody models that are compact and portable and thus well-suited for robotic applications.

7. Sebastian Wrede, COR-Lab, University of Bielefeld, Germany,
Title: A Domain-specific Language and Simulation Architecture for the Oncilla Robot.
We present an experimentation environment for the exploration of rich motor skills on a quadruped robot. We introduce a simulation platform around the (open) quadruped robot ’Oncilla’ that is integrated with a domain-specific language (DSL) tool-chain for modeling and execution of reproducible movement experiments. The Webots-based simulator, features an open source and cross-platform (Linux and Mac OS) interface with taxonomic structure and rich sensor feedback. By providing a common abstracted interface for both simulation and hardware, it enables faster and easier transfer of experiments between these two domains. This interface is accessible through a local C++ interface as well as remotely via an open-source middleware in with C++, Java, Python, and Common Lisp bindings for extended language and tool support. Integrated with the common interface, a DSL-based environment around the simulator allows compact formulation of experiments in domain-specific terminology and eased generation of reproducible artifacts for experimentation. We present how movement architectures and dynamical-systemsbased components of different partners of the AMARSi project are incorporated in the development of the DSL and the experimentation toolchain.

8. Andrea Del Prete, Instituto Italiano di Tecnologia, Italy,
Title: Software Tools for Dynamics, Simulation, Identification, Estimation and Control: the Open-Source iCub Project.
The new generation of force-controlled robots was born to safely and effectively interact with humans and the environment. The need for force control affects various aspects of robotics, starting from robot modeling, which moves from a kinematic to a dynamic level. Fast and reliable computation of robots dynamics is now paramount, because it impacts many other aspects, such as: state estimation, inertial parameter identification, dynamic simulation, inverse dynamics control. In this paper we present the results of our efforts to deal with robots dynamics and related aspects. All the presented software tools are written in C++ and they are available in the opensource iCub repository, which is released under a GPL license.

9. Jonas Buchli, ETH, Switzerland,
Title: Software Tools for Modeling, Simulation and Real-Time Control of Legged Robots in Complex Environments.
In this talk we will review our experience on using simulators to prototype control software for fast legged robots in real complex environments. In such environments the robots have to work robustly, with certain performance and safety guarantees. We look at the software from the perspective of the final goal; that is having a robot performing in real world. This goal determines the specifications, requirements and properties for the methods and the tool chain that is employed. Some of the requirements that arise, like efficiency, real time capability seem to be, at first sight, in conflict with our desire to have general purpose and user friendly tools and well maintainable code. Requirements like safety and correctness seem to be at odds with fast prototyping and flexible software.

10. Houman Dallali, Instituto Italiano di Tecnologia, Italy,
Title: Development of a Dynamic Simulator for Compliant Humanoid Robot Based on a Symbolic Multibody Approach.
An open source, dynamic simulator for the COmpliant huMANoid robot, COMAN is presented. The key advantages of this simulator are: it generates efficient symbolic dynamical equations of the robot with high degrees of freedom, it includes a user-defined model of the actuator dynamics (the passive elasticity and the DC motor equations), user defined ground models and fall detection. Users have the freedom to choose the proposed features or include their own models. The models are generated in Matlab and C languages, where the user can leverage the power of Matlab and Simulink to carry out analysis also have the flexibility of C language for real-time experiments on DSP. This simulator can be downloaded from the IIT website.

11. Christian Duriez (University of Lille),
Title: Control of Elastic Soft Robots Using Real-Time Inverse Simulation on SOFA Framework.
In this work, we propose a new method for the control of soft robots with elastic behavior, piloted by several actuators. The approach is based on the real-time computation of the Finite Element Method (FEM) using the framework SOFA. Using an optimization algorithm, based on a reduced compliance matrix, the model is inverted in real-time. It is used in a control loop, to find the contribution of the actuators (force and/or position) that deforms the structure so that the terminal end of the robot follows a given position. The optimization integrates the internal characteristics of the actuators and the constitutive law of the deformable structure. It is also coupled with the collision response pipeline of SOFA so additional constraints, like rigid or deformable obstacles are integrated in the control algorithm. We illustrate our method using simulated examples of both serial and parallel structures and we validate it on a real 3D soft robot made of silicone.

12.Vojtech Vonasek (Czech Technical University),
Title: Sim: A Light-weight Robot Simulator for Modular Robots.
Physical simulation is frequently used in robotics, as it allows evaluation of control strategies or planning techniques in situations, where experiments with real robots are time consuming or limited due to power resources or reliability issues. Such an example is locomotion generation for modular robots formed from many small robotic modules. To find a suitable locomotion for such a complex robot, evolutionary principles can be used, which requires to conduct many experiments. To speed up this process, a robot simulator can be employed. In this paper, a novel, light-weight open-source simulator for the modular robots is introduced. It provides both physical and sensor simulation and it was designed to be run in a headless mode, i.e., without any visualization. This allows to run it on e.g. on computational grids. Despite this fact, the progress of the simulation can be later visualized using external tools like Blender 3D. This brings advantage in comparison to more general and powerful simulators that cannot be easily run on such machines. The paper briefly introduces architecture of the simulator with description of its utilization in evolutionary modular robotics.

13. Antonio Morales & Beatriz Leon (University Jaume I),
Title: A Framework for Robot and Human Grasping Simulation.
The human hand and its dexterity in grasping and manipulating objects are some of the hallmarks of the human species. However, although many researchers have partially studied various aspects, to date there has been no comprehensive characterization of the human hands function for grasping and manipulation of everyday life objects. Our hypothesis is that the confluence of both scientific fields, the biomechanical study of the human hand and the analysis of robotic manipulation of objects, greatly benefits and advances both disciplines. Additionally, we believe that the use of a simulation framework in which we could model and validate each of the processes involved in dexterous grasping is crucially important. Therefore, in this work, the current knowledge of robotics and biomechanics guides the design and implementation of a simulation framework focused on manipulation interactions that allows the study of the grasp through simulation.

14. Davide Piovesan (Northwestern University),
Title: Haptic Simulator of Abnormal Biomechanics During Patient-Clinician Interaction.
We present the development of a haptic simulator of abnormal limb biomechanics. The system can be used as a training device for clinicians to understand the causality of the haptic sensation associated with neurological impairments and abnormal biomechanics during manipulation of a patients’ limb. Other uses include the testing of operational research algorithms to maximize the exchange of haptic information; reproducing conditions of human-human interaction; and a test bed for developing novel assessment techniques and targeted interventions. Experimental results support the hypothesis that the best assessment of abnormalities can be obtained via specific manipulation strategies.


March 15th, 2013: Submission deadline for Abstracts


March 30th, 2013: Notification of acceptance.

May 10th, 2013: Workshop date


Time Talk
8:30  - 9:00    Houman Dallali (IIT), Welcome, introduction & Talk on A Dynamic Simulator for the Compliant Humanoid Robot, COMAN.

9:00  - 9:30 Nathan Koenig (OSRF), The many faces of simulation: Use cases for a general purpose simulator.

9:30  - 10:00  Abhi Jain (JPL), Structure Based Modeling and Computational Architecture for Robotic Systems.

10:00 - 10:30

Coffee Break

10:30 - 10:55 Shin'ichiro Nakaoka (AIST), Choreonoid: Extensible GUI Framework for Handling Virtual Robots.

10:55 - 11:20

Christian Duriez (INRIA), Control of Elastic Soft Robots Using Real-Time Inverse Simulation on SOFA Framework.

11:20 - 11:45 Paul Fisette (CEREM, UCL), ROBOTRAN, Self-contained Multibody Models for Robot Modeling and Real-Time Control.

11:45 - 12:10

Marc Freese (Coppelia Robotics), Distributed Control in Robotics Simulations: a Case Study.

12:10 - 12:35

Sebastian Wrede (Uni. Bielefeld), A Domain-specific Language and Simulation Architecture for the Oncilla Robot.

12:45 - 14:00

Lunch Break

14:00 - 14:20 Oskar von Stryk & Katayon Radkhah (TU Darmstadt), Dynamic Modeling for Simulation, Optimization & Control of Bio-Inspired Elastic Robots.

14:20 - 14:45 Jonas Buchli (ETH), Software Tools for Modeling, Simulation and Real-Time Control of Legged Robots in Complex Environments.

14:45 - 15:05

Andrea Del Prete (IIT), Software Tools for Dynamics, Simulation, Identification, Estimation and Control: the Open-Source iCub Project.
15:05 - 15:25

Vojtech Vonasek (Czech Technical Uni.), Sim: A Light-weight Robot Simulator for Modular Robots.
15:30 - 16:00
Coffee Break

16:00 - 16:20

Antonio Morales & Beatriz Leon (Uni. Jaume), A Framework for Robot and Human Grasping Simulation.

16:20 - 16:40

Davide Piovesan (Northwestern Uni.), Haptic Simulator of Abnormal Biomechanics During Patient-Clinician Interaction.

16:40 - 17:15

Time for discussion, conclusion.


Full proceedings are available here


This workshop is supported by the FP7 European Project AMARSI ( ).


[1] Online, 2013, IEEE Spectrum: DARPA awards simulation software contract to open source robotics foundation, URL:

[2] Online, 2013, OpenHRP Dynamic Simulator, URL:

Last Updated on Wednesday, 17 August 2016 14:12


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