Optimized machine vision setups are a requirement for precise and efficient product quality assurance. As the design space is high-dimensional, a manual design requires a lot of engineering experience and experimental work, associated with high costs and often non-optimal results. For automatic evaluation, there are evaluation metrics such as measurement uncertainty and scan resolution to evaluate the quality of an inspection. However, it is not trivial how to combine different criteria to optimize the setup based on the inspection requirements. We propose to fuse the metrics through a probabilistic surface inference to quantify the amount of information gained by a specific setup configuration. To this end, the product surface is modeled by a random process and the problem is adapted to a Gaussian Process (GP) inference. We introduce a local inference based on the local surface orientation and propose a novel spectrum-based approach to determine the GP parameters based on the required inspection resolution. The inference results, based on simulations, are demonstrated for the inspection of a cylinder head.

Most computer vision applications are built from a combination of basic computer vision algorithms, such as filters, descriptors and matchers. The functionality of this computer vision toolbox is well understood and solid implementations exist. One of the leading and most often used implementations is the Open Source Computer Vision Library (OpenCV), which implements more than 500 computer vision algorithms. However, most of these algorithms have multiple parameters that have to be tuned to the specific vision application. Therefore, during development of new vision applications, some human-in-the-loop iteration cannot be avoided. The problem is that OpenCV lacks support for this phase of development. This usually leads to a situation where each developer creates his own static ad-hoc debug solutions that output intermediate results in a printf-debugging manner. To remedy this situation, we developed reusable general-purpose interactive tools for visual debugging and development of OpenCV-based computer vision applications.

In this paper we present a new autonomous mobile robot for research focused on the field of biotechnology and miniaturized automation. The robot “Microman”, itself is modular structured by a basic and functional unit. With a diameter of 50mm and an omni-directional driving system the robot is able to move in comparison to other systems in a very limited area. Due to the wireless charging system in combination with a variety of different contactless sensors the use in contaminated applications is possible. In addition the robot comes with a versatile arena with several application modules, which are described in the system overview. Here a cell screening application is selected. The mechanical and electrical design as well as the novel color-based navigation system and first results complete this paper.

We present an approach to track a point cloud with a 3D camera system with low latency and/or high frame rate, based on ground truth provided by a second 3D camera system with higher latency and/or lower frame rate. In particular, we employ human tracking based on Kinect cameras and combine it with higher frame-rate/lower latency of Time-of-Flight (ToF) cameras. We present the system setup, methods used and evaluation results showing a very high accuracy in combination with a latency reduction of up to factor 30.

Modular robots are built of many basic robotic modules that can be connected into robots of various shapes. These robots are able to recover from a failure by ejecting and replacing damaged modules. Although this type of failure recovery is usually suggested in literature, it may be precluded due inability of the broken modules to cooperate or if no spare modules are available. In such a case, locomotion of a damaged robot should be adapted to allow the robot to reach a repair station or even to finish its task without the need to exchange the broken modules. In this paper, we investigate how to adapt motions of modular robots with respect to failures using the concept of motion planning with motion primitives. The ability of the proposed system to recover from failures is verified in a simulation and also in a HW experiment.

In this paper we present a novel approach for intuitive telemanipulation in Cartesian space and discuss the results of a user study evaluating different aspects of our approach. The proposed method inhibits certain degrees of freedom based on the current viewpoint. Together with automatic mapping of the input device to corresponding motion axes, our approach provides a very intuitive method for controlling the telemanipulation system while reducing the mental workload of the operator and therefore the amount of erroneous commands. Similar principles apply for controlling the viewpoint of real or virtual cameras to facilitate manipulation or navigation tasks.

In this paper we present applications for a robot system whose gripper is equipped with distributed capacitive tactile proximity sensors (CTPS). Firstly, we introduce and evaluate a closed loop control scheme by which it is possible to align the gripper to objects or features of the environment using proximity values alone. We call this control method proximity servoing. It is implemented by equilibrating the sensor signals, resulting in a robust preshape in all 6DOF of the object pose. Objects can then be grasped with virtually no displacement. Secondly, also based on proximity servoing, we demonstrate and evaluate novel ideas for combined haptic and proximity-based exploration. Without any cues from an external camera the system is capable of detecting and exploring features such as curvatures, edges or corners of objects. It is also shown that tactile and proximity exploration steps can be used complementarily to increase efficiency in exploration while delivering accurate object measurements.

Achieving seamless human-robot cooperation requires the robot to recognize its surroundings and be aware of human actions and their implications. Still, no recognition method, e.g., for objects or actions, can achieve zero failure rates. Thus, robotic actions might be selected and planned based on false assumptions. This inherits risks for humans and machinery. Consequently, the verification of plausibility of recognition results considered in context can extract inconsistencies and regulate robotic behavior due to adapted risk assessment. In this paper, an approach for plausibility verification in a situation awareness system and corresponding risk estimation adaptation is presented and examined.

The integration of artificial skin into robotic systems has long been foreseen. For the last decades the introduction was always just a couple of years away. One of the possible reasons for the slow progress is the focus of the research efforts on high sensitivity and spatial resolution. The resulting tactile sensor prototypes are presented as laboratory prototypes with no or little chance to be successfully integrated into a robotic system. In order to enable the operation of an artificial skin on a robotic system in real-world applications the conflict of goals between high sensitivity and an overload-proof, collision tolerant, design needs to be solved. Within this paper we present a stretchable tactile surface sensor as a major functional component of an artificial skin setup that is able to unite the required sensitivity with the ability to withstand collisions. The results of first experiments regarding sensitivity and the ability of the DLR Artificial Skin setup to withstand high impact forces are presented.

In this paper we base upon capacitive tactile proximity sensor modules developed in a previous work to demonstrate applications for safe human-robot-interaction. Arranged as a matrix, the modules can be used to model events in the near proximity of the robot surface, closing the near field perception gap in robotics. The central application investigated here is object tracking. Several results are shown: the tracking of two human hands as well as the handling of occlusions and the prediction of collision for object trajectories. These results are important for novel pretouch- and touch-based humanrobot interaction strategies and for assessing and implementing safety capabilities with these sensor systems.

The endoscopic guidance system for minimally invasive surgery presented here autonomously aligns the laparoscopic camera with the end-effectors of the surgeon's instruments. It collects information on the movements of the instruments from former interventions and can therefore predict them for autonomous guidance of the endoscopic camera. Knowledge is extracted by trajectory clustering, maximum likelihood classification and a Markov model to predict states. Alternative movements in an ongoing intervention are modeled. A first prototype of a robotic platform for minimally invasive surgery is described, which has two instrument arms, an autonomous robotic camera assistant and two haptic devices to control the instrument arms. The approach of long-term prediction and optimal camera positioning was tested in a phantom experiment with a hit rate of over 89% for predicting the movement of the end-effectors. Including this prediction for computing the camera position, leads to 29.2% less movements and to an improved visibility of the instruments.

This paper presents a novel approach for haptic object recognition with an anthropomorphic robot hand. Firstly, passive degrees of freedom are introduced to the tactile sensor system of the robot hand. This allows the planar tactile sensor patches to optimally adjust themselves to the object's surface and to acquire additional sensor information for shape reconstruction. Secondly, this paper presents an approach to classify an object directly from the haptic sensor data acquired by a palpation sequence with the robot hand - without building a 3d-model of the object. Therefore, a finite set of essential finger positions and tactile contact patterns are identified which can be used to describe a single palpation step. A palpation sequence can then be merged into a simple statistical description of the object and finally be classified. The proposed approach for haptic object recognition and the new tactile sensor system are evaluated with an anthropomorphic robot hand.

This paper presents an algorithm to detect finger collisions in Multi-fingered robot hand SDH-2. The need for this feature is discussed and advantages are shown. It will be presented, in which way the algorithm is build into a basic development environment of reactive grasping and what are possible problems and possibilities in collaboration with the tactile sensor elements.

This paper outlines a method for applying a kinematic path following control of a mobile robot without any regard for surface structure. Background. A great deal of mobile robotics kinematics analysis is based on the movement of the robot on a two dimensional flat surface. Our application for precision surgery required a new approach to a system that would operate on a highly non-linear surface; this specific system was a surgical robot that would conduct craniotomies while moving over highly irregular and often deformed skulls. Methods. The approach used an abstract view of the operating environment that would totally ignore the surface, instead determining the control parameters based only on the robot and the desired cutting trajectory. The approach was then evaluated in a 3D environment using a series of predefined surfaces to determine bounding limits in the control theory. These limits were then tested in a second series of tests using real data from the CT preoperative imagery of previous patients and phantoms. The simulation results were then compared with the actual performance of the robot on phantoms and cadavers. Results. The approach has been successfully implemented on the first medical robot to position itself through spiked wheels on the surface of the skull. Testing has to date been successful in both a simulation environment, and on initial phantom and cadiever trials, with accuracies equal to that of the larger industrial modified surgical robots.

This paper describes the content of the video for ICRA 2010. The approach presented in the video is a complete automatic registration of all needed devices for robotic guided CO2 bone processing. The system consists of an optical tracking system, a lightweight robot and a scan head. While a standard point to point transformation for the transformation between optical tracking system and robot is inadequate, a special registration algorithm was developed and therefore is used. This allows an automatic registration by moving the robot to different positions and storing the position of a body attached to the robot TCP. The scan head is registered to the optical tracking system or to the robot using a camera that tracks the prototype laser. Only the patient must be registered manually. The result is that the robot moves to the target position on the bone defined inside the segmented 3D CT dataset. With the known registration between robot and tracking system the robot can also be controlled via visual servoing.

A new autonomous mobile robot for research, education and Arts is presented. The robot features a rich variety of sensors and an extendable processing unit while still maintaining very small size (approx. 51 mm in diameter). The robot can be used in combination with a very versatile arena for robot experiments that allows to setup and execute experiments in an easy and automated fashion, which also includes automatic recharging of the robot. We will provide an overview of the basic modular hardware design of the robot including a detailed description of the cpu, power management, bus system, actuators and sensors. A special section will be dedicated to a detailed description of the communication capabilities of the robot. Furthermore we will present the software architecture which provides high-level access to the robots hardware.

In order to enable robots to execute highly dynamic tasks in dangerous or remote environments, a semiautomatic teleoperation concept has been developed and will be presented in this paper. It relies on a modular software architecture, which allows intuitive control over the robot and compensates latency-based risks by using Augmented Reality techniques together with path prediction and collision avoidance to provide the remote user with visual feedback about the tasks and skills that will be executed. Based on this architecture different skills with high dynamics are integrated in the robot control, so that they can be executed autonomously without the delayed feedback of the user. The skill-based grasping by adherence of smooth or fragile objects during a remote controlled picking and placing task will be exemplary presented.

In this paper we propose a new diversity measure based on the correlation of bit strings for the analysis of genetic programming (GP) experiments. The diversity measure has been applied to analyse the impact of pruning on the diversity of a population during genetic programming and its relation to the convergence time of the fitness function. To show the usability of the proposed diversity measure a GP experiment is introduced where simulated Jasmine robots have to learn a collision avoidance behaviour to find their way through a maze. A full analysis of this experiment is given with different fixed pruning strategies in respect to the population diversity and fitness. The GP has been done on behaviour-based robot controllers implemented in MDL2isin. MDL2isin has the advantage that it provides a very compact bit string representation of the control programme, which can be used for diversity analysis.

Material processing using a laser has become a widely used method for industrial procedures (e.g. laser welding or cutting). Furthermore the medical laser has become an integral part of dermatology, neurosurgery, ENT, esthetic, plastic and general surgery. Recent publications have shown, that the short pulsed CO2laser is suitable to ablate bony and cartilage tissue and proposes fundamentally new operative techniques in medicine. The obtainable precision in cutting (in the hundred micrometers range) with a laser system can only be reached using means of computer and robot assisted surgery. We established the first robot assisted laser bone ablation setup, comprising a prototype CO2laser system and a six degree of freedom robot. The laser beam is guided through a passive articulated mirror arm to the robots end-effector. The end-effector is composed of a two mirror galvanometric scan head, which deflects the pulsed laser beam onto the tissue. In this paper we present an end-effector calibration and registration method to determine the parameters which are critical in obtaining precise and accurate cutting results.

In this paper, a tactile sensing system for an anthropomorphic robot hand is presented. The tactile sensing system is designed as a construction kit making it very versatile. The sensor data preprocessing is embedded into the hand's hardware structure and is fully integrated. The sensor system is able to gather tactile pressure profiles and to measure vibrations in the sensor's cover. Additionally to the introduction of the hardware, the signal processing and the classification of the acquired sensor data will be explained in detail. These algorithms make the tactile sensing system capable to detect contact points, to classify contact patterns and to detect slip conditions during object manipulation and grasping.

This paper focuses on the simplification of the on-line programming process of industrial robots. It presents in detail the input part of a modular on-line programming environment presented as overview in [1]. Main concept of this programming environment is an intuitive way of moving and teaching robots, while supporting the user with assisting algorithms like collision avoidance and automatic path planning. Goal is the combination of different approaches from tele-operation, programming by demonstration, Virtual Reality and off-line programming, and to reuse them in a new fashion on-line on the shop-floor. This paper presents some ideas and concepts, how input devices and strategies for robot programming could look like and how to use them to set up an intuitive manual motion control of the robot.

In this paper, we present the very first virtual reality training system for neurosurgical interventions based on a real surgical microscope and on a haptic interface for a better visual and ergonomic realism. Its main purpose is the realistic simulation of the palpation of low-grade glioma. The ability of a surgeon to feel the difference in consistency between tumors cells and normal brain parenchyma requires considerable experience and it is a key factor for a successful intervention. The simulation takes advantage of an accurate tissue modeling, a force feedback device and the rendering of the virtual scene directly to the oculars of the operating microscope.

Robot assisted interventions often require coupling and decoupling of the robot to/from a specific tool. By using manual gripper changing systems these operations are facilitated, but the robot has to approach to and move away from the coupling position. Industrial applications are mostly based on movements which are teached-in, since the working environment is perfectly described (i.e. working cell). Especially in robot assisted surgery we are facing non fixed tools to which the robot has to be coupled (e.g. a holding device attached to a mobilised bone) and restricted working areas with special safety requirements. In this paper we present an automatic end-effector registration method and a semiautomatic coupling procedure exemplarily for robot assisted orthognathic surgery. By using means of an optical localisation system and force- /torque sensing, the coupling procedure is controlled by a multi- sensor data fusion approach. The developed methods can be adapted to any robot assisted intervention.

This paper focuses on the simplification of the on-line programming process of industrial robots. It presents a modular on-line programming environment (software and hardware), which supports an intuitive way of moving and teaching robots, while supporting the user with assisting algorithms like collision avoidance and automatic path planning. Main idea is the combination of different approaches from tele-operation, programming by demonstration and off-line programming, and reuse them in a new fashion on-line on the shopfloor. The proposed programming environment is designed to evaluate the usability of different combination of assisting techniques.

This paper presents a multi-sensor based generic approach to opening doors for a dexterous robot. Once the handle has been located by a computer vision algorithm and properly grasped, we are able to open doors without using a model or other prior knowledge of the door geometry. This is done by combining the sensor information of both a force-torque sensor in the robot wrist and a tactile sensor matrix in the robot gripper itself. Our experimental results show that the combination of both sensors achieves the most successful way to open the door.

We present a comparative study of two spatially resolved macroscopic models of an autonomous robotic swarm. In previous experiments, the collective behavior of 15 autonomous swarm robots, driven by a simple bio-inspired control algorithm, was investigated: in two different environmental conditions, the ability of the robots to aggregate below a light source was tested. Distinct approaches to predict the dynamics of the spatial distribution were made by two different modeling approaches: one model was constructed in a compartmental manner (ODEs). In parallel, a space-continuous model (PDEs) was constructed. Both models show a high degree of similarity concerning the modeling of concrete environmental factors (light), but due to their different basic approaches, show also significant differences in their implementation. However, the predictions of both models compare well to the observed behavior of the robotic swarm, thus both models can be used to develop further extensions of the algorithm as well as different experimental setups without the need to run extensive real robotic preliminary experiments.

The construction of kinematics models by hand is often a tedious and error-prone process. This paper presents an overview of the Organic Robot Control project. One main goal of the project is to automate the computation of closed-formed solutions for both the forward and inverse kinematics of general industrial manipulators. For simple manipulator geometries, we use rule-based programming techniques to obtain a closed-form solution. For complicated geometries, we use elimination techniques based on the computation of Groebner basis. In so doing we convert a complicated coupled transcendental set of kinematics equations into a univariate polynomial. This polynomial can much easier be solved. The generated analytic models fulfill real time requirements and can be parallelized very well.

This paper presents a tactile language for controlling a robot through its artificial skin. This language greatly improves the multimodal human-robot communication by adding both redundant and inherently new ways of robot control through the tactile mode. We defined an interface for arbitrary tactile sensors, implemented a symbol recognition for multi-finger contacts, and integrated that together with a freely available character recognition software into an easy-to-extend system for tactile language processing that can also incorporate and process data from non-tactile interfaces. The recognized tactile symbols allow for both a direct control of the robot's tool center point as well as abstract commands like "stop" or "grasp object x with grasp type y". In addition to this versatility, the symbols are also extremely expressive since multiple parameters like direction, distance, and speed can be decoded from a single human finger stroke. Furthermore, our efficient symbol recognition implementation achieves real-time performance while being platform-independent. We have successfully used both a multi-touch finger pad and our artificial robot skin as tactile interfaces. We evaluated our tactile language system by measuring its symbol and angle recognition performance, and the results are promising.

An open-loop method on the basis of acceleration compensation to reduce liquid surface oscillation generated during a high-speed transfer will be presented in this paper. In order to suppress the undesirable liquid vibration effects, so called `sloshing', a new simple and effective methodology consisting of adapting the gripper orientation is proposed. The assumption of slosh-free movement will be valid so long as there is no relative motion between the container and the liquid. To accomplish this objective, the maximum acceleration in every time-instant has to be considered in the computation. This represents that our method operates basically in maintaining the normal of the liquid surface opposite to the entire systems acceleration until the completion of the transportation. Experimental results using a manipulator KUKA-KR16 will be demonstrated to validate the effectiveness of our approach.

During the rapid transfer of freely suspended loads, hazardous swing oscillations may be produced. In order to deal with this problem, we propose a new methodology, which derives immediately from the acceleration compensation principle[1]. It consists basically of modifying the original reference-trajectory comprised by a set of suspension points, with the purpose to compensate undesirable residual swing effects at the end of the transfer motion. This open-loop method is computationally simple, time-efficient and feasible. In contrast to the popular methods utilized to control the overhead cranes, which traveling motions are restricted only in the horizontal plane, our compensation technique involves as well the motion in vertical plane. The rope length is assumed invariant. A set of simulations and experimental verifications using an industrial robot manipulator KUKA KR 16 have been carried out to demonstrate the feasibility of the proposed solution.

Intuitive human-robot cooperation presents a challenge to robots since it demands of them a high level of understanding of the human user. Our approach is to estimate the human intention and select and proactively execute an appropriate robot task without requiring an explicit user command. This paper describes the concept and details of our proactive execution module and how it forms an integral part of our intuitive human-robot cooperation system. We present implementation details and the results of an evaluation scenario.

Two control algorithms for a swarm robot are presented that enable it to orientate itself by using information from the geometry of trail bifurcations within a trail network. The development of these algorithms was inspired by the behavior of Pharaoh's ants as reported by Jackson et al., 2004. The performance of the robot is analyzed in a large number of embodied experiments with different bifurcation angles. The reactive behavior implemented by simple rules is sufficient to accomplish this task using a robot of limited capabilities. The frequency of correct reorientations is maximized when the trail bifurcation angle is 60 degrees, as found in natural networks

Micro robots in large scale swarms often have a very restricted program memory which limits the robot's application range. We present a finite state machine operating system for swarm micro robots, that can overcome such problems and gives the designer of swarm algorithms a tool that is easy to handle. The operating system's flow control or rather the robot's control program is represented in the Motion Description Language Two Extended (MDL2epsiv). MDL2epsiv is based on MDLepsiv but has been extended to a fully functional behaviour description language as shown in this paper. The MDL2epsiv based control programs are encoded in a byte code that is interpreted on a micro controller. The byte-code concept significantly reduces the size of the control program which will be shown in this paper.

Today's palletizing and handling robots move their payloads on trajectories that are neither optimized for minimal grasping forces nor for careful object handling. Problems arise with vacuum grippers that can only exert normal forces on payloads. During high-speed transfers undesired shear forces and torques appear that may well lead to the loss of grasped objects and consequently to a production stop and damage to both the object and the environment. A related problem is concerned with handling goods that need special care, such as liquids that should not spill over during transport, or bakery goods on a tray that should not be scrambled and dislocated before being moved into an oven. To be able to handle objects with care but nevertheless with maximum velocity, a new simple and effective methodology based on the adaptation of the gripper orientation is proposed. It leads to robot trajectories (in particular gripper orientations) that minimize shear forces and torques on grasped objects and thus allows gentle robotic handling processes. The methodology is presented as well as simulation and experimental results

This paper describes the initial work of the EC-funded project I-SWARM which aims to take a leap forward in robotics research, in distributed and adaptive systems as well as in understanding self-organizing biological swarm systems. The next step from today's micro robotics research will be the mass-production of micro robots, which can then be employed as a "real" swarm consisting of up to 1,000 robot clients. These clients will all be equipped with limited, pre-rational on-board intelligence. The swarm will consist of a huge number of heterogeneous robots, differing in the type of sensors, manipulators and computational power or behaviours and programs. Such a robot swarm can be employed for a variety of applications, including micro assembly nano-handling, biological, medical or cleaning tasks

This article describes the proposal for an architecture that allows to easily connect different devices or modules to a system for computer integrated surgery (CIS). This would allow the reuse of expensive CIS devices like robots. Yet, in contrast to other surgical robot systems, this architecture does not require a robot - it just supports its use. Therefore, a robot and its attached tool are distinguished as two separate devices. All devices in this architecture have knowledge about themselves and thus hardly depend on other devices. The article motivates and describes this architecture and explains an example implementation.

In this paper we present a new method for the cooperative carrying of a rigid object by a robot and a human. This method is based on tactile sensor input. In the carrying mode the robotic system defines constraints for the user's possible movements when manually guiding the object (and robot arm). Thus, instead of using a complex recognition process for interpreting the user's intention the robot determines the next action on basis of the defined constraints and the measured tactile sensor data. The defined constraints only allow the human partner to manipulate the carried object during the carrying procedure in the manner of a water hand pump. In this way, so-called in-place-translations of the object in all directions are possible, while the human partner maintains full control over orientation and position of the object.

The application of robots in the same workspace with humans results, intended or unintended, in direct mechanical interaction. This requires additional sensory abilities of the robots. Besides sensor systems that help the robots to structure their environment, like cameras, radar sensors, etc., a sensor system on the robot's surface is needed that is able to detect mechanical contacts of the robot with its environment. Because of shadings it is not possible to exclude potential collisions in spite of the "structuring" sensor systems. Therefore a tactile sensor system is essential for reasons of security but also as support of the robot control system and additional communication channel. In this paper we propose the requirements to such a system related to its application on an autonomous humanoid robot. According to these postulated requirements a prototypal system was built and integrated in an existing experimental setup which consists of a redundant robot arm with 7 degrees of freedom in a human-similar kinematics. The sensor and its technical specifications as well as the experimental setup are described in the second part of this paper.

The presented mobile microrobots are employed inside the vacuum chamber of a scanning electron microscope (SEM). Very often more than one robot is required even for simple handling tasks due to the unfamiliar force ratios in the micro world. This paper describes how the SEM is used as a position sensor system that is a presupposition of the automatic coordination of microrobots. For depth measurements a triangulation principle with the help of the electron beam is used. First results and the required calibration methods are presented.

The ultimate ambition of a robotic system in surgical theater has to be the safety of the involved humans: patient, physicians, nurses. In order to provide that claim the criteria of ergonomics, redundancy, short reaction time, and accuracy must be met. In contrast to many other surgical robotic applications we are studying complex bone cut trajectories which require several distinct orientations of the robot tool and milling cutter, respectively. Exact bone cuts are especially needed for bone repositionings placed at the human skull in craniofacial surgery. Therefore, the system has to be flexible enough to keep the capability of changing the patient's position during the intra-operative phase. This paper introduces the overall concept and realization of the system.

In the scanning electron microscope, many studies require micromanipulation tools that can be flexibly and comfortably used for the handling and positioning of objects ranging from a few millimeters down to fractions of microns. In the paper, some aspects of the control of small mobile microrobots by teleoperation are addressed facing the special behavior of their piezoelectric actuators and the challenge of handling tasks in ranges of very different orders of magnitude (100 nm-100 mm). In order to give the teleoperated tweezers a kind of low-level "intelligence", a control loop using the SEM as a sensor system is being developed. In different open-loop handling experiments, the employed 6D-mouse proves satisfactory as an intuitive user interface.

A new algorithm for segmenting and tracking deformable (shape-changing) linear (thin and long with negligible diameter) objects in video images is presented. The core of the algorithm is to detect the two boundary curves of a deformable linear object within adaptive tracking windows. The boundaries are found by analyzing the gray value gradients within the tracking windows, The algorithm provides a robust tracking of the shape- and width-changing object by using a fuzzy estimation of the object's membership function at the tracked points. The result of the segmentation is a list of base describing the shape of the object. This list is used to define features for shape change detection. For the feature analysis, we propose an algorithm which works well for several classes of feature curves. Experiments on robotic applications using both algorithms show promising results.

A new algorithm for segmenting and tracking deformable (shape-changing) linear (thin and long with negligible diameter) objects in video images is presented. The core of algorithm is to detect the two boundary curves of a deformable linear object within adaptive tracking windows. The boundaries are found by analyzing the gray value gradients within the tracking windows. The algorithm allows moderate changes in the image brightness by introducing tolerance parameters to the algorithm. It also provides a robust tracking of the shape- and width-changing object by using a fuzzy estimation of the object's membership function at the tracked points. The result describes the shape of the object and is used to define features for shape change detection.

Grasping of unmodelled objects in unstructured scenes is a significant task in numerous robot applications. In order to reach short cycle times, grasp planning must make the best of the image information acquired all at once, which comprises more or less visual occlusions and thus unknown regions. For time reasons, it is not acceptable to inspect in further steps all unknown regions by means of local sensing before grasp planning. We present a grasp planning system, which determines explicitly the fractions of grasp operations in known and unknown regions. Using that crucial information, a planning system may not request a tool change, but prefer a grasping operation with the mounted gripper, which moves mostly in known and free regions. Then, only a small part has to be executed slowly using local sensing with, for examples, proximity or force torque sensors.

Presents a new operation planning system which enables a surgeon to plan and perform complex robot-supported surgical interventions. The system has been evaluated in the Clinic for Cranio-Maxillo-Facial Surgery at the University of Heidelberg. In contrast to commercial systems, our goal was that the system should consider the complete surgical intervention and not just a single procedure of it. A second goal was that the system should enable the management of complex operations, independent of which way the intervention is intra-operatively performed (without technical support, with passive navigation support or with active support by robots). Our system supports the surgeon during pre-operative planning as well as during the intra-operative execution phase. Therefore, we developed a model of the course of operation by which the management of surgical interventions is made possible. The focus of this paper is on this course model. First, we introduce instruction graphs and describe the structure of each activity, observing its attributes and their context. Additionally, various surgical scopes are presented which enable the surgeon to select one view among different ones in the individual operation procedures, in accordance with medical and technical knowledge as well as in accordance with different degrees of abstraction. Finally, we demonstrate the realisation of the introduced course model in our operation planning system and present first results in clinical practice.

In this paper, a flexible manipulation system used in a catalogue sale business is presented. One application of such a system is the object turning operation for the barcode search. The regrasping procedures are studied in detail to its significance in object turning. This work focuses on regrasping tasks with several different grippers.

This paper deals with the robot handling of deformable linear objects (DLO), such as hoses, wires or leaf springs, based on contact state transitions and force/moment sensing. First, we describe the characteristic change of force and moment for all single-contact transitions which may occur between the DLO and a rigid polyhedron obstacle. Second, we discuss the sensor data processing for the automated detection of the state transitions.

Deals with the robust manipulation of deformable linear objects such as hoses or wires. We propose manipulation based on the qualitative contact state between the deformable workpiece and a rigid environment. First, we give an enumeration of possible contact states and discuss the main characteristics of each state. Second, we investigate the transitions which are possible between the contact states and derive criteria and conditions for each of them. Finally, we apply the concept of contact states and state transitions to the description of a typical assembly task.

Presents an approach to parallel path planning for industrial robot arms with six degrees of freedom in an online given 3D environment. The method is based a best-first search algorithm and needs no essential off-line computations. The algorithm works in an implicitly discrete configuration space. Collisions are detected in the Cartesian workspace by hierarchical distance computation based on polyhedral models of the robot and the obstacles. By decomposing the 6D configuration space into hypercubes and cyclically mapping them onto multiple processing units, a good load distribution can be achieved. We have implemented the parallel path planner on a workstation cluster with 9 PCs and tested the planner for several benchmark environments. With optimal discretisation, the new approach usually shows very good speedups. In online provided environments with static obstacles, the parallel planning times are only a few seconds.

The paper is based on a path planning approach for industrial robot arms with 6 degrees of freedom in an online given 3D environment. It has online capabilities by searching in an implicit and discrete configuration space and detecting collisions in the Cartesian workspace by distance computation based on the given CAD model. Here, we present different methods for specifying the C-space discretization. Besides the usual uniform and heuristic discretization, we investigate two versions of an optimal discretization for a user-predefined Cartesian resolution. The different methods are experimentally evaluated. Additionally, we provide a set of 3-dimensional benchmark problems for a fair comparison of the path planner. For each benchmark, the run-times of our planner are between only 3 and 100 seconds on a Pentium PC with 133 MHz.