Malte Rast

Developing Virtual Testbeds for mobile robotic applications in the woods and on the moon

Simulation in the context of engineering often focuses on very special details of global systems. Robot designers usually begin with the analysis of new actuators and joint designs. This corresponds to a “bottom-up”-strategy in the development of simulation models. For classical fields of application of robotics, e.g. in production plants with a well defined environment this is the approved method, because it allows very detailed insights into the analyzed subsystems. On the other hand, unpredictable effects of the interaction of multiple subsystems may easily be overseen. In particular, nontechnical environments like in moon exploration tasks or in a biological environment like in forestry applications are hard to describe in an analytical way to integrate them into an analytical simulation model. This is why this paper presents the idea and some practical aspects of the development of “Virtual Testbeds”. In a Virtual Testbed, the entire system is simulated as a whole in Virtual Reality - not only small subsystems of a global system. According to the requirements different subsystems are simulated with different levels of detail. In contrast to the classical “bottom-up”-strategy this can be seen as a “top-down”-approach. Therefore the employment of a multi-body dynamics system as a platform for the development of versatile simulation and testing environments is proposed. Using the examples of the evaluation and testing of an extraterrestrial walking exploration robot design and the development of a method for self-localization in forestry, the idea is further deepened. As a special field of attention the integration of a method of soil simulation as a particular requirement of a Virtual Testbed for walking exploration robots is presented.

Mental Models for Intelligent Systems: eRobotics Enables New Approaches to Simulation-Based AI

AbstracteRobotics is a newly evolving branch of e-Systems engineering, providing tools to support the whole life cycle of robotic applications by means of electronic media. With the eRobotics methodology, the target system and its environment can be modeled, validated, and calibrated to achieve a close-to-reality simulation. In this contribution, we present simulation-based mental models for autonomous systems as a foundation for new approaches to prediction and artificial intelligence. We formulate a methodology to construct optimization problems within simulation environments in order to assist autonomous systems in action planning. We illustrate the usefulness and performance of this approach through various examples in different fields. As application for space robotics, we focus on climbing strategies of a legged mobile exploration robot. Furthermore, we enable skillfull interaction control in service robotics and address energy consumption issues. The contribution concludes with a detailed discussion of the concept presented here.

Close to Reality Simulation of Bulk Solids Using a Kind of 3D Cellular Automaton

The development of algorithms providing a close to reality simulation of dynamic virtual worlds made substantial technological progresses during the last decade — contrary to the close to reality simulation of bulk solids. Standard simulation methods like particle or rigid-body simulation are not applicable to this simulation problem because a large number of elements is needed for convincing simulation results which cannot be handled in real-time. In this paper we present a kind of 3-dimensional cellular automaton which can handle a large number of elements at the cost of a spatial discretization. This approach is combined with state of the art rigid body simulation techniques resulting in a close to reality simulation of bulk solids in real-time applications.Copyright

A Virtual Testbed for Human-Robot Interaction

Many research efforts in human-robot interaction (HRI) have so far focussed on the mechanical design of intrinsically safe robots, as well as impedance control for tasks in which human and robot will work together. However, comparatively little attention is paid to an approach that ensures safety and permits close HRI while dispensing human with the necessity of physical presence in close proximity to the robot. In this work we acquire real human manipulation gestures using Microsofts Kinect sensor and project them in realtime into the workspace of a simulated, impedance controlled robot manipulator. This way, we can remotely and therefore safely superimpose human motion over the robots dynamic motion, wherever the human operator is located. The simulated robot state is then transferred to the real robot as input, so as to physically perform the intended task. The Virtual Testbed approach might not only be useful for HRI pre-analysis, testing and validation goals but particularly advantageous for telepresence, industrial and harzardous tasks as well as training purposes. Simulation results are provided to show the effectiveness of the approach.

Simulation-Based Engineering with Hybrid Testbeds

The development of new software applications in mobile robotics is very challenging due to the complexity oftasks and the multitude of involved components. The hybridtestbed idea, combining the advantages of virtual and realtestbeds, shortens the development cycle with little effort.Hybrid testbeds provide the possibility to exchange real andsimulated components at any time without influencing thefurther data processing pipeline. In this paper, we describea robust communication model based on generic interfacesallowing the transfer of data and commands across system borders. Based on this generic interface concept, hybrid testbeds, containing simulated as well as real hardware components, can easily be set up. Furthermore, the presented concept additionally allows physically separated systems, and thus supports simulation-based engineering combining development strategies like hardware-in-the-loop, co-design and concurrent engineering. The applicability of the introduced approach is shown in the development process of a mobile localization unit. It consists of several sensors and an internal processing unit for position estimation. Starting in a virtual environment, the unit has been implemented by replacing the simulated components successively by their physical counterparts leading to a final setup with real hardware. Finally, the simulation system itself is used to command the localization unit and to visualize the raw sensor and processed data in the virtual environment.

Fast VR Application Development Based on Versatile Rigid Multi-Body Dynamics Simulation

More and more areas in research and development use Virtual Reality technologies. To quickly realize new applications at low costs, the reuse of existing functionality is of high importance. In the area of mobile robotics, physics based simulation components promise optimal reusability: The physical laws always stay the same and do not depend on the application. Hence, as long as the applications try to emulate reality, physics based simulation software will be reusable. Unfortunately, depending on the kind of application, different simulation models for different physical domains are needed: Particle models for fluids and soil, finite-elements for non-rigid bodies, multi-body systems and so on. However, for those applications developed at Institute for Man-Machine Interaction at the RWTH Aachen University, a multi-body dynamics component has taken a central role in the process of application development. It is fully integrated within a modern 3D-simulation and visualization tool. It is enhanced by generalized tools of contact graph analysis, which support the fast development of robust applications suitable for daily use. The paper discusses the benefit of this multi-body system as a platform for versatile application development, taking the following three applications as examples: The first example is the development of forest machine simulators for usage in education and training of machine operators. The existence of a purely kinematically realized, phenomenological implementation with widely equivalent range of functions allows a direct comparison of the programming efforts. The second example is the development of algorithms for space robot motion planning. The example demonstrates, how easy and effective innovative robotic simulation applications can be realized using a common, dynamics based simulation framework. The third example finally describes the development of a Virtual Testbed for legged lunar exploration robots. The Virtual Testbed example handles in detail the concept of “top-down-development” of simulation models. The refinement of the simulation of foot-soil-contact situations using a force exchange interface and the refinement of the actuator dynamics simulation using an energy exchange interface serve as examples.Copyright

Distributed Information Processing and Rendering for 3D Simulation Applications

The growing demand for 3D simulation techniques in various application domains leads to more and more specialized tools and complex frameworks. Between homogeneous or inhomogeneous clients, data has to be distributed and synchronized in centralized or decentralized setups. Hardware/Software-in-the-Loop and Co-Simulation are common tasks in virtual prototyping. Load balancing and parallelization is necessary for computationally intensive simulations. Spatially distributed developers and designers collaborate in networked virtual environments. All these different applications impose different requirements on the data distribution and synchronization mechanism. In this paper, we categorize distribution scenarios, their requirements and according synchronization techniques. Four different approaches with different key aspects are presented and compared by means of a reference implementation and several application examples. This overview shall enable the reader to choose the approach best suited for his particular distribution problem.

Combining 3D Simulation Technology With Object-Oriented Databases: A Database Oriented Approach to Virtual Reality Systems

Using object-oriented databases as the primary data source in VR applications has a variety of advantages, but requires the development of new techniques concerning data modeling, data handling and data transfer from a Virtual Reality system’s point of view. The many advantages are outlined in the first part of this paper. We first introduce versioning and collaboration techniques as our main motivation. These can also be used in the traditional file based approach, but are much more powerful when realized with a database on an object and attribute level. Using an object-oriented approach to data modeling, objects of the real world can be modeled more intuitively by defining appropriate classes with their relevant attributes. Furthermore, databases can function as central communication hubs for consistent multi user interaction. Besides, the use of databases with open interface standards allows to easily cooperate with other applications such as modeling tools and other data generators. The second part of this paper focuses on our approach to seamlessly integrate such databases in Virtual Reality systems. For this we developed an object-oriented internal graph database and linked it to object-oriented external databases for central storage and collaboration. Object classes defined by XML data schemata allow to easily integrate new data models in VR applications at run-time. A fully transparent database layer in the simulation system makes it easy to interchange the external database. We present the basic structure of our simulation graph database, as well as the mechanisms which are used to transparently map data and meta-data from the external database to the simulation database. To show the validity and flexibility of our approach selected applications realized with our simulation system so far e. g. applications based on geoinformation databases such as forest inventory systems and city models, applications in the field of distributed control and simulation of assembly lines or database-driven virtual testbeds applications for automatic map generation in planetary landing missions are introduced.© 2011 ASME

Developing Virtual Testbeds for Tasks in Research and Engineering

Simulation in the context of robotic engineering often focuses on very special details of global systems. For example robot designers usually begin with the analysis of new actuators and joint designs. This corresponds to a “bottom-up”-strategy in the development of complex simulation models. This is probably a good choice for classical fields of robotic applications, e.g. in production plants with well defined system states and environments, because it allows very detailed insights into the analyzed subsystems. On the other hand, unpredictable effects of the interaction of multiple subsystems may easily be overseen. To overcome this problem, this paper presents the idea and some practical aspects of the implementation of a virtual test environment (Virtual Testbed). In a Virtual Testbed, the entire system is simulated as a whole in virtual reality — not only small subsystems of a global system. According to requirements simulation of special subsystems is refined by specific simulation methods and integrated into the overall simulation framework. In contrast to the classical “bottom-up”-strategy this can be seen as a “top-down”-approach in the development of complex simulation models. Therefore a platform for the development of versatile simulation and testing environments is presented. Using the example of the evaluation and testing of an extraterrestrial walking exploration robot design in its Virtual Testbed, the idea is further deepened. As a special field of attention the integration of a method of soil simulation for the refinement of foot-ground-interaction as a particular requirement of this kind of simulation is described.Copyright

Architecture for Integrated Development of Complex Systems With Model-Based System Specification and Simulation

R&D projects in space robotics are characterized by the profound integration of various disciplines and the associated complexity of the systems to be designed. To support this multidisciplinary collaboration, the paradigm of Model-Based Systems Engineering (MBSE) is a suitable approach. Another essential challenge of space projects is the early verification and validation of the model-based design. Tests with physical prototypes give rise to considerable costs and can reproduce the real conditions only to a certain extent.In this paper, we present an approach to integrate the cross-domain system specification with the cross-domain test design with Virtual Testbeds to a comprehensive system model. This enables a continuous simulation-based engineering in the design of space applications. Requirements from the early design phase can be used systematically to test and integrate detailed models from the various disciplines.The presented system model offers the potential to efficiently integrate national and international know-how to support the development phases of a mission from the requirements analysis to design, visualization, control, operation, training up to marketing and technology transfer. Future projects can be realized faster, less expensive and more robust.Copyright