Refine
Year of publication
Document Type
- Conference Proceeding (20)
- Article (8)
- Book (2)
- Part of a Book (2)
- Contribution to a Periodical (1)
- Report (1)
Is part of the Bibliography
- no (34)
Institute
- Fachbereich 1 - Institut Informatik (34) (remove)
The investigation of neuronal accounts of cognition is closely linked to collaboration between behavioral experiments, theory and application and supports the process of moving from pure behaviorist correlation analysis to gaining a real understanding of the underlying mechanisms. Cognition builds upon the individual behavioral history, and the understanding of cognition is based on neuronal principles.
The study of human behavior incorporates in particular interactive, dynamically changing scenarios with multiple human individuals. Both the acquisition of behavioral data of human subjects, the modeling of behavior, as well as the evaluation in interactive scenarios, makes it necessary to generate simulated images of reality. Simulations allow the investigator to precisely control the structure of the environment the subject interacts with. Furthermore, situations that would be too dangerous in the real world (e.g. near-crash driving situations) can be investigated using virtual reality.
By nature, simulated reality frameworks are designed to simulate naturalistic environments. Within these environments, ecologically relevant stimuli embedded in a meaningful and controlled context can be presented. The quality of experimental data acquired within the simulated environment depends not to the last on the degree of immersion of the human subject.
Driving experiments usually attempt to relate observable driver behavior to cognitive inputs. The precise visual (retinal) input of a driver in a driving simulator depends also on the exact position of his head with respect to the screen (Noth et al., 2010). The major meaning of ego motion feedback can be considered as a continuous calibration here.
In a virtual cooperation scenario, consistency matters - if an operator perceives an object at 1 m distance, moving 20 cm towards it should decrease the perceived distance to 80 cm, moving to the side of an object which occludes another one should reveal the latter (Pretto et al., 2009).
The ego-motion feedback mitigates the cues that remind operators of the fact that they are in a virtual and not in the real world. The way the appearance of a virtual object changes due to a lateral head movement is identical to its real counterpart, which means that even relations between real and virtual objects remain (Creem-Regehr et al., 2005; Cutting, 1997).
In this contribution we introduce a head tracking system which is utilized to incorporate human ego motion in simulated environments improving immersion in the context of a human-robot collaborative task and in an interactive driving simulator.
For both cases, we explain how the ego motion feedback leads to a more precise comprehension of the virtual scene and how the aspect of immersion influences the feeling of being “really” inside of the virtual scene and the weakening of the awareness of the border between the real and the virtual world.
The neuronal basis of movement preparation, during which movement parameters such as movement direction are assigned values, is fairly well understood (Georgopoulos, 2000). Motor and premotor cortex as well as portions of the parietal cortex represent movement parameters through the activity of neuronal populations (Bastian et al., 2003; Cisek & Kalaska, 2005).
The parameter representation is of dynamic nature, updated in the course of movement. It adapts to boundary conditions of the motion plan or to environmental changes. Schwartz (2004) was able to decode motor cortical activity in the motor cortex and utilized this knowledge to drive a virtual or robotic end-effector. Thus he proved that the motor cortex is involved in the generation of movement planning. At this level of abstraction we assume that the movement of an end-effector, as well as human walking movement, is represented appropriately by its direction and satisfies other constraints, such as obstacle avoidance or movement coordination.
A neuronal dynamic of movement generates goal-directed movements and satisfies other constraints, such as obstacle avoidance. Movement is generated by choosing low-dimensional, behaviorally relevant state variables. Behavioral goals are represented as attractors of dynamical systems over such behavioral variables (Schöner et al., 1995). The robots trajectory emerges as a solution of these dynamical systems, in which the behavioral variables are stabilized at attractors corresponding to behavioral goals. Constraints are included in a similar manner as repellers. Recently we applied this approach to generate reaching movements for manipulators under obstacle avoidance and orientation con- straints (Iossifidis & Schöner, 2009; Reimann et al., 2010a,b).
We aim to develop an approach to robotic action based on dynamical systems 1
that is quantitatively modeled on human behavior. By varying the intrinsic parameters obtained for different individuals we will be able to implement different personal styles of movement. In this contribution we implement the neuronal dynamics of movement on a humanoid robotic system which generates goal-directed walking movements while avoiding obstacles.
Advances in Human-Robot Interaction" provides a unique collection of recent research in human-robot interaction. It covers the basic important research areas ranging from multi-modal interfaces, interpretation, interaction, learning, or motion coordination to topics such as physical interaction, systems, and architectures. The book addresses key issues of human-robot interaction concerned with perception, modelling, control, planning and cognition, covering a wide spectrum of applications. This includes interaction and communication with robots in manufacturing environments and the collaboration and co-existence with assistive robots in domestic environments. Among the presented examples are a robotic bartender, a new programming paradigm for a cleaning robot, or an approach to interactive teaching of a robot assistant in manufacturing environment. This carefully edited book reports on contributions from leading German academic institutions and industrial companies brought together within MORPHA, a 4 year project on interaction and communication between humans and anthropomorphic robot assistants.
As service robotics research advances rapidly, availability of objective, reproducible test specifications and evaluation criteria and also of benchmarking is more and more felt to be desirable in the community. As a first step towards benchmarking, in this paper we propose a formalization of tests - exemplified for domestic grasp&place tasks. The underlying philosophy of our approach is to confront the robot system in a black-box manner with requirements of a “rational customer”, and characterize the performance of the system in an objective way by the outcomes of a test-suite tailored to this scenario. A formalized single test description consists of a clear and reproducible specification of the robot’s task and the full context on the one hand, and a number of figures which objectively characterize the test result on the other hand. We illustrate this methodology for the domestic assistance scenario.
CoRA is a robotic assistant whose task is to collaborate with a human operator on simple manipulation or handling tasks. Its sensory channels comprising vision, audition, haptics, and force sensing are used to extract perceptual information about speech, gestures and gaze of the operator, and object recognition. The anthropomorphic robot arm makes goal-directed movements to pick up and hand-over objects. The human operator may mechanically interact with the arm by pushing it away (haptics) or by taking an object out of the robot’s gripper (force sensing). The design objective has been to exploit the human operator’s intuition by modeling the mechanical structure, the senses, and the behaviors of the assistant on human anatomy, human perception, and human motor behavior.
To enable a robotic assistant to autonomously reach for and transport objects while avoiding obstacles we have generalized the attractor dynamics approach established for vehicles to trajectory formation in robot arms. This approach is able to deal with the time-varying environments that occur when a human operator moves in a shared workspace. Stable fixed-points (attractors) for the heading direction of the end-effector shift during movement and are being tracked by the system. This enables the attractor dynamics aoproach to avoid the spurious states that hamper potential field methods. Separating planning and control computationally, the approach is also simpler to implement. The stability properties of the movement plan enable to approach to deal with fluctuating and imprecise sensory information. We implement this approach on a seven degree of freedom anthropomorphic arm reaching for objects on a working surface. We use an exact solution of the inverse kinematics, which enables us to steer the spatial position of the elbow clear of obstacles. The straight-line trajectories of the end-effector that result far from obstacles make the movement goals of the robotic assistant predictable, improving man-machine interaction.
Das übergeordnete Forschungsgebiet, in das sich die vorliegende Arbeit einbettet, befasst sich mit der Erforschung von informationsverabeitenden Prozessen im Gehirn und der Anwendung der resultierenden Erkenntnisse auf technische Systeme. In Analogie zu biologischen Systemen, deren Beschaffenheit aus den Anforderungen der Umwelt an ihr Verhalten resultiert, leitet sich die Anthropomorphie als Entwurfsprinzip für die Struktur des mit den Menschen interagierenden robotischen Assistenzsystemen ab. Der Autor behandelt in der vorliegende Arbeit das Problem der Erzeugung von Motorverhalten im dreidimensionalen Raum am Beispiel eines anthropomorphen Roboterarmes in einem anthropomorphen robotischen Assistenzsystem. Entwickelt wurde hierbei ein allgemeiner Ansatz, der die Konzepte der Erzeugung von Motorverhalten im 3D-Raum, der Voraussimulation dynamischer Systeme zur Systemdiagnose und zur Suche gewünschter Systemzustände, sowie ein Konzept der Organisation von Verhalten enthält und vereinigt. Nichtlineare dynamische Systeme bilden das mathematische Fundament, die einheitlich, formale Sprache des Ansatzes, mit der sowohl das Motorverhalten des Roboters als auch dessen zeitkontinuierliche Teilsysteme rückgekoppelt werden.
Auf gute Zusammenarbeit
(2008)
Simulated reality environment incorporating humans and physically plausible behaving robots, providing natural interaction channels, with the option to link simulator to real perception and motion, is gaining importance for the development of cognitive, intuitive interacting and collaborating robotic systems. In the present work we introduce a head tracking system which is utilized to incorporate human ego motion in simulated environment improving immersion in the context of human-robot collaborative tasks.
Autonomous robots with limited computational capacity call for control approaches that generate meaningful, goal-directed behavior without using a large amount of resources. The attractor dynamics approach to movement generation is a framework that links sensor data to motor commands via coupled dynamical systems that have attractors at behaviorally desired states. The low computational demands leave enough system resources for higher level function like forming a sequence of local goals to reach a distant one. The comparatively high performance of local behavior generation allows the global planning to be relatively simple. In the present paper, we apply this approach to generate walking trajectories for a small humanoid robot, the Aldebaran Nao, that are goal-directed and avoid obstacles. The sensor information is a single camera in the head of the robot. The limited field of vision is compensated by head movements. The design of the dynamical system for motion generation and the choice of state variable makes a computationally expensive scene representation or local map building unnecessary.
In the presented work we compare machine learning techniques in the context of lane change behavior performed by humans in a semi-naturalistic simulated environment. We evaluate different learning approaches using differing feature combinations in order to identify appropriate feature, best feature combination, and the most appropriate machine learning technique for the described task. Based on the data acquired from human drivers in the traffic simulator NISYS TRS 1 , we trained a recurrent neural network, a feed forward neural network and a set of support vector machines. In the followed test drives the system was able to predict lane changes up to 1.5 sec in beforehand.
To enable a robotic assistant to autonomously reach for and transport objects while avoiding obstacles we have generalized the attractor dynamics approach established for vehicles to trajectory formation in robot arms. This approach is able to deal with the time-varying environments that occur when a human operator moves in a shared workspace. Stable fixed points (attractors) for the heading direction of the end-effector shift during movement and are being tracked by the system. This enables the attractor dynamics approach to avoid the spurious states that hamper potential field methods. Separating planning and control computationally, the approach is also simpler to implement. The stability properties of the movement plan make it possible to deal with fluctuating and imprecise sensory information. We implement this approach on a seven degree of freedom anthropomorphic arm reaching for objects on a working surface. We use an exact solution of the inverse kinematics, which enables us to steer the spatial position of the elbow clear of obstacles. The straight-line trajectories of the end-effector that emerge as long as the arm is far from obstacles make the movement goals of the robotic assistant predictable for the human operator, improving man-machine interaction
For any kind of assistant systems, the ability to interact with the human operator and taking into account his or her assumptions and expectations, is the basis for a reasonable behavior. As a consequence the human behavior have to be studied in order to generate driver models that are learned from human driving data. In this work we focus on the improvement of the immersion in driving simulation environment by developing and implementing a cheap and efficient method for head tracking. We also explain why head tracking feedback is crucial for the quality of collected behavioural data, especially for simulators with close screen distances.
In this paper we describe an architecture for behavioral organization based on dynamical systems. This architecture
enables the generation of complex behavioral sequences, which is demonstrated using the example of approaching and
passing a door. The behavioral sequence is generated by activating and deactivating the elementary behaviors dependent
on sensory information and internal logical conditions. The architecture is demonstrated on a mobile KOALA robot and
in simulation as well.
CORA is a robotic assistant whose task is to collaborate with a human operator on simple manipulation or handling tasks. Its sensory channels comprising vision, audition, haptics, and force sensing are used to extract perceptual information about speech, gestures and gaze of the operator, and object recognition. The anthropomorphic robot arm makes goal-directed movements to pick up and hand over objects. The human operator may mechanically interact with the arm by pushing it away (haptics) or by taking an object out of the robot's gripper (force sensing). The design objective has been to exploit the human operator's intuition by modeling the mechanical structure, the senses, and the behaviors of the assistant on human anatomy, human perception, and human motor behavior.
We extend the attractor dynamics approach to generate goal-directed movement of a redundant, anthropomorphic arm while avoiding dynamic obstacles and respecting joint limits. To make the robot's movements human-like, we generate approximately straight-line trajectories by using two heading direction angles of the tool-point quite analogously to how movement is represented in the primate central nervous system. Two additional angles control the tool's spatial orientation so that it follows the tool-point's collision-free path. A fifth equation governs the redundancy angle, which controls the elevation of the elbow so as to avoid obstacles and respect joint limits. These variables make it possible to generate movement while sitting in an attractor (or, in the language of the potential field approach, in a minimum). We demonstrate the approach on an assistant robot, which interacts with human users in a shared workspace
The presented work formulates an framework in which early prediction of drivers lane change behavior is realized. We aim to build a representation of drivers lane change behavior in order to recognize and to predict driver's intentions as a first step towards a realistic driver model. In the test bed of the Institute of Neuroinformatik, based on the traffic simulator NISYS TRS 1, 10 individuals have driven in the experiments and they performed more then 150 lane change maneuvers. Lane-offset, distance to the front car and time to contact, were recorded. The acquired data was used to train - in parallel- a recurrent neural network, a feed forward neural network and a set of support vector machines. In the followed test drives the system was able of performing a lane change prediction time of 1.5 sec beforehand. The proposed approach describes a framework for lane-change detection and prediction, which will serve as a prerequisite for a successful driver model.
Generating collision free reaching movements for redundant manipulators using dynamical systems
(2010)
For autonomous robots to manipulate objects in unknown environments, they must be able to move their arms without colliding with nearby objects, other agents or humans. The simultaneous avoidance of multiple obstacles in real time by all link segments of a manipulator is still a hard task both in practice and in theory. We present a systematic scheme for the generation of collision free movements for redundant manipulators in scenes with arbitrarily many obstacles. Based on the dynamical systems approach to robotics, constraints are formulated as contributions to a dynamical system that erect attractors for targets and repellors for obstacles. These contributions are formulated in terms of variables relevant to each constraint and then transformed into vector fields over the manipulator joint velocity vector as an embedding space in which all constraints are simultaneously observed. We demonstrate the feasibility of the approach by implementing it on a real anthropomorphic 8-degrees-of-freedom redundant manipulator. In addition, performance is characterized by detecting failures in a systematic simulation experiment in randomized scenes with varying numbers of obstacles.
Generating flexible collision-free reaching move-
ments is a standard task for autonomous articulated robots that
is critical especially when such systems interact with humans in
a service robotics setting. Current solutions are still challenging
to put into practice. Here we generalize an approach
first
used to plan end-effector movement that is based on attractor
dynamical systems. We show, how different contributions to
the motion planning dynamics can be formulated in constraint-
specific reference frames and then transformed into the frame
of the joint velocity vector. We implement this system on an
8 DoF redundant manipulator and show its feasibility in a
simulation. A systematic experiment with randomly generated
obstacle scenes characterizes the performance of the system.
Especially challenging confi
gurations of obstacles are discussed
to illustrate how the method solves these cases
Integrating Orientation Constraints into the Attractor Dynamics Approach for Autonomous Manipulation
(2010)
We present an architecture based on the Dynamic Field Theory for the problem of scene representation. At the core of this architecture are three-dimensional neural fields linking feature to spatial information. These three-dimensional fields are coupled to lower-dimensional fields that provide both a close link to the sensory surface and a close link to motor behavior. We highlight the updating mechanism of this architecture, both when a single object is selected and followed by the robot's head in smooth pursuit and in multi-item tracking when several items move simultaneously
Temporal stabilization of discrete movement in variable environments: An attractor dynamics approach
(2009)
The ability to generate discrete movement with distinct and stable time courses is important for interaction scenarios both between different robots and with human partners, for catching and interception tasks, and for timed action sequences. In dynamic environments, where trajectories are evolving online, this is not a trivial task. The dynamical systems approach to robotics provides a framework for robust incorporation of fluctuating sensor information, but control of movement time is usually restricted to rhythmic motion and realized through stable limit cycles. The present work uses a Hopf oscillator to produce discrete motion and formulates an online adaptation rule to stabilize total movement time against a wide range of disturbances. This is integrated into a dynamical systems framework for the sequencing of movement phases and for directional navigation, using 2D-planar motion as an example. The approach is demonstrated on a Khepera mobile unit in order to show its reliability even when depending on low-level sensor information.
Based on the concepts of dynamic field theory (DFT), we present an architecture that autonomously generates scene representations by controlling gaze and attention, creating visual objects in the foreground, tracking objects, reading them into working memory, and taking into account their visibility. At the core of this architecture are three-dimensional dynamic neural fields (DNFs) that link feature to spatial information. These three-dimensional fields couple into lower dimensional fields, which provide the links to the sensory surface and to the motor systems. We discuss how DNFs can be used as building blocks for cognitive architectures, characterize the critical bifurcations in DNFs, as well as the possible coupling structures among DNFs. In a series of robotic experiments, we demonstrate how the DNF architecture provides the core functionalities of a scene representation.
We describe the general concept, system architecture, hardware, and the behavioral abilities of Cora (Cooperative Robot Assistant, see Fig. 1), an autonomous non mobile robot assistant. Outgoing from our basic assumption that the behavior to perform determines the internal and external structure of the behaving system, we have designed Cora anthropomorphic to allow for humanlike behavioral strategies in solving complex tasks. Although Cora was built as a prototype of a service robot system to assist a human partner in industrial assembly tasks, we will show that Cora’s behavioral abilities are also conferrable in a household environment. After the description of the hardware platform and the basic concepts of our approach, we present some experimental results by means of an assembly task.