Dominik Henrich

University of Bayreuth · Institute for Computer Science

Natural language allows robot programming to be accessible to everyone. However, the inherent fuzziness in natural language poses challenges for inflexible, traditional robot systems. We focus on instructions with fuzzy time requirements (e.g., "start in a few minutes"). Building on previous robotics research, we introduce fuzzy skills. These defin...

Human-robot collaboration enables highly adaptive co-working. The variety of resulting workflows makes it difficult to measure metrics as, e.g. makespans or idle times for multiple systems and tasks in a comparable manner. This issue can be addressed with virtual commissioning, where arbitrary numbers of non-deterministic human-robot workflows in a...

Various automation tasks require the precise specification of forces/torques over time and space to be applied to the environment. Traditionally, only experts can provide such specifications, contrasting the need for intuitive robot programming in small and medium-sized enterprises. This paper presents a concept for extending kinesthetic robot prog...

Human-robot teaming is crucial for future automation in small and medium enterprises. In that context, domain-specific process models are used as an intuitive description of work to share between two agents. Process designers usually introduce a certain degree of abstraction into the models. This way, models are better to trace for humans, and a si...

The ability to review robot programs before they are executed can be used to correct erroneous programming. In complex processes, such a review can only be achieved by integration of additional peripheral devices and workpieces used into the programming framework. In this work, we present a semi-automatic method for the calibration of a turntable a...

For robot manipulators, it is nowadays necessary to know their surroundings. This knowledge consists at least of a world representation with recognized objects. During the reconstruction of scene objects from multiple views, changes, like positioning of the objects, or additional unwanted signals, like parts of a human co-worker, may occur. In this...

Human-robot teaming receives an ever-increasing level of attention in research, development and industry. Novel approaches to task sharing in hybrid teams range from optimized schedules to intelligent cobot assistants with a high degree of autonomy. These approaches must prove their usefulness and benefits compared to manual work or full automation...

Opening up untapped potential in smaller enterprises requires methods for robot programming usable by non-experts. In a previous work on an automata-based programming approach without a graphical interface, linking to previous states to close loop structures was identified as challenging for users. In this paper, we propose an approach to generate...

The optimization of robot programs before their execution enables to correct inaccurate and erroneous programming. In this work, an extension to kinesthetic programming is presented, with the aim to optimize robot trajectories in their geometrical and temporal course. For this, we define a set of optimization methods, which can be used quickly and...

Playback robot programming is fast and easy to use for non-experts, because the robot only needs to be manually guided. However, it is only capable of replaying the trajectory exactly as it was taught. We present the concept of loop increments for playback programmed robots to allow the user to teach tasks like palletizing or stacking without havin...

Robot-based automation is still not widespread in small and medium-sized enterprises, since programming industrial robots is usually costly and only feasible by experts. This disadvantages can be resolved by using intuitive robot programming approaches like playback programming. At the same time, there are currently not automatized automatized, lik...

Virtual commissioning can be used for planning human-robot workflows. To this end, digital human models enable consideration of human ergonomics and motion times as optimization criteria. This leads to optimal but fixed human-robot schedules with clearly quantified benefits over the manual process. By contrast, recent research on robot co-workers h...

Eine zunehmende Variantenvielfalt durch kundenindividuelle Produkte führt zu besonderen Anforderungen an die Logistik und die Montage in Produktionsumgebungen. Komponenten und Werkzeuge müssen flexibel und auftragsspezifisch an der entsprechenden Station bereitgestellt und genutzt werden. Diesem Sachverhalt stehen hohe und weiter steigende Löhne in...

Automation in smaller enterprises requires methods for robot programming usable by non-experts. In this paper, we propose an approach based on finite state automata. The basic principle is kinesthetically guiding the robot through tasks and programming state transitions to represent absolute movement, relative movement (to the current pose or to th...

We can only allow human-robot-cooperation in a common work cell if the human integrity is guaranteed. A surveillance system with multiple cameras can detect collisions without contact to the human collaborator. A failure safe system needs to optimally cover the important areas of the robot work cell with safety overlap. We propose an efficient algo...

A representation of the world as a 3D model is a common necessity in robotics and automation. In previous work, we developed a concept to generate boundary representation (B-Rep) models from multiple point clouds using a hand-held depth-camera and to register them without a prior known pose. During the online reconstruction, properties of the senso...

Human-Robot Teams offer the flexibility needed for partial automation in small and medium-sized enterprises (SMEs). They will thus be an integral part of Factories of the Future. Our research targets a particularly flexible teaming mode, where agents share tasks dynamically. Such approaches require cognitive robots with reasoning and sensing capabi...

A central part of a robot programming system is its user interface. Oftentimes, its design and development are neglected, although the user interface is the connection between the robot and the user. In this work we present the application of a user-centered design process to a robot programming system interface. The iterative process consists of t...

Recent research in robotics envisions shared human-robot workspaces to combine individual advantages of humans and robots. As part of this vision, robot manipulators must avoid collisions with humans and other a priori unknown obstacles in the shared workspace. State-of-art approaches (e.g. certainty grids, bounding volume hierarchies) test robot p...

Flexible human-robot cooperation can be coordinated by dynamic task allocation, where team members observe task progress and repeatedly decide to execute outstanding operations from a commonly known task model. Such task-level coordination strategies do not cover cases, where a high level of abstraction for task modelling implies operations infeasi...

This paper describes an approach to intuitive robot programming, with the aim of enabling non-experts to generate sensor-based, structured programs. The core idea is to generate a variant of a finite state automaton (representing the program) by kinesthetic programming (physically guiding the robot). We use the structure of the automaton for contro...

We propose a highly flexible approach to human-robot cooperation, where a robot dynamically selects operations contributing to a shared goal from a given task model. Therefore, knowledge on the task progress is extracted from a world model constructed from eye-in-hand camera images. Data generated from such partial workspace observations is not rel...

Programming industrial robots is usually costly, tedious, and only feasible by experts. Hence, these systems do not play a big role in small and medium sized enterprises. In this paper, we present a novel approach for designing a robot programming system with the playback programming paradigm inspired by video editing concepts. First of all, we pre...

Manually guiding a robot to a predefined goal position is nearly impossible, because the user needs to align the tool center point of the robot precisely with the predefined position. In this paper we present an approach to enable guiding a robot to a goal position with the help of multi-modal feedback in form of haptic feedback by increasing the s...

Human-robot cooperation is a trending topic in robotics research. Recent works mainly focus the execution phase, where human and robot actions need to be planned and coordinated based on given task models. In contrast, we discuss an approach to fast creation of task models suitable for flexible human-robot teaming: Inspired by graphical robot appli...

General purpose robots are established tools in a variety of industrial applications. An important goal in actual research is to transfer the advantages of these tools into more unstructured environments like househoulds or small and medium sized enterprises. One challenge in this field is to enable non-experts to use all the capabilities of a robo...

Three dimensional models are represented in various ways. One possibility are boundary representation (B-Rep) models, which contain geometric and topological information. This makes B-Reps suitable for tasks that need an explicit algebraic representation of the surface, e.g. numerical optimization or simulation. Reconstructing a B-Rep model of an o...

Usually, programming robot systems is expensive and complex and only profitable for companies with big lot sizes. Hence these systems do currently not play a big role in small or medium sized enterprises. This work extends the programming paradigm published in [1] that is based on the playback programming method, so that also sensor information can...

One long term goal of artificial intelligence and robotics research is the development of robot systems, which have approximately the same cognitive, communicational, and handling abilities like humans. This yields several challenges for future robot systems. For instance in the field of communicational abilities, future robot systems have to bridg...

One long term goal of artificial intelligence research is the development of robot systems, which have approximately the same cognitive, communicational, and handling abilities like humans. For instance, a human-like communication requires an intuitive and symbolic user interface. Since actual robot control systems typically consist of subsymbolic...

A graphics processing unit (GPU)-based implementation of a space carving method for the reconstruction of the photo hull is presented. In particular, the generalized voxel coloring with item buffer approach is transferred to the GPU. The fast computation on the GPU is realized by an incrementally calculated standard deviation within the likelihood...

Controlling and programming multi-arm robots is cost-intensive and complex. As a result, such systems currently do not play a big role for small and medium sized enterprises. This paper tackles the task of intuitive multi-robot playback programming and introduces a new programming paradigm that allows teaching of both, single arm and multi-arm robo...

Boundary representation (B-Rep) models are surface representations containing geometric and topological information that are used in many fields such as reverse engineering, simulation and optimization. To create a B-Rep model from point clouds of different viewpoints, the data are usually merged into one global point cloud and subsequently a surfa...

Actual robot systems are typically used within an industrial environment, where they perform tasks in the areas of manufacturing, assembly, or inspection. Many of these tasks require the definition and execution of sensor-based motions. Such motions are typically defined using a subsymbolic representation like manipulation primitive nets [1] or iTa...

Human-robot interaction in private households demands easy and natural interfaces, as well as an understanding of areas of application. Two fundamental issues are multimodal interaction (i.e., multifarious ways of communication with the robot) and task allocation (i.e., division of labour between the user and the robot). We report on an explorative...

One long term goal of robotic research is the development of robot systems, which have approximately the same cognitive, communicational, and manipulative abilities like humans and which can be operated by non-experts. The usage by non-experts sets additional requirements to the abilities of a robot system. For instance, the system should be able t...

A robot is essentially a real-time, distributed embedded system operating in a physical environment. Often, control and communication paths within the system are tightly coupled to the actual hardware configuration of the robot. Furthermore, the domain contains a high amount of variability on different levels, ranging from hardware, over software t...

To provide robots for a wide range of users, there needs to be an easy and intuitive way to program them. This issue is addressed by the robot programming by demonstration paradigm, where the user demonstrates the task to the robot. While there exist a lot approaches that use multiple demonstrations for the learning procedure, single-shot robot pro...

In order to provide robots for a wide range of users, they must be easily programmable. This is addressed by various approaches of programming a robot by demonstration, where the user guides a robot through the task. We propose an approach that adapts a once demonstrated trajectory to a new situation. The system extracts the relevant reference fram...

We present a GPU-based implementation of the photo hull using the generalized voxel coloring with item buffer (GVC-IB) approach with the aim of reconstructing humans online in surveillance scenarios. To allow for a fast computation on the GPU, an incremental calculated standard deviation is used in the likelihood ratio test that is applied as color...

Human-robot cooperation requires efficient and precise geometric reconstruction of objects in any shared workspace to meet safety requirements. State-of-art solutions accept a trade-off to satisfy these requirements. Usually, this implies limited robot speed, expensive hardware, or low-precision input and output. In contrast, we present a novel mul...

A long-term goal in current robotic research is the development of intuitive interfaces for human-robot interaction. Here, one field of application are object manipulation tasks. Such tasks consist of grasping, moving, and placing objects [1]. In this work, we focus on the subtask of moving an object, which is also called the handling of the object...

A Personal Robot should be able to handle unknown objects in unknown environments. For a manipulation task the question of what to do with an object once it has been grasped, is one of the most essential ones besides the grasping task itself. Moreover, the planning time should be at least as fast as the time the robot needs for its grasping/placing...

A Personal Robot should be able to handle un-known objects in unknown environments. For a manipulation task the question what to do with an object once it had been grasped is one of the most essential ones beside the grasping task itself. Moreover, the planning time should be at least as fast as the time the robot needs for its motions. We propose...

Programming robots is a complicated and time-consuming task. A robot is essentially a real-time, distributed embedded system. Often, control and communication paths within the system are tightly coupled to the actual physical configuration of the robot. Thus, programming a robot is a very challenging task for domain experts who do not have a dedica...

In this paper an approach is introduced to human robot interaction in a known scenario with unknown human intentions. Initially, the robot reacts by copying the human action. As the human-robot interaction proceeds, the level of human-robot interaction improves. Before each reaction, the robot hypothesizes its potential actions and selects one that...

Behavior-based robotic manipulators are very flexible since they can perform many different tasks without reprogramming. Unfortunately none of the existing approaches is able to interweave multiple manipulation tasks and execute them reliably at the same time, enabeling e.g. an intuitive human-robot cooperation. To bridge this gap, we suggest a nov...

Programming robots is a complicated and time-consuming task. A robot is essentially a real-time, distributed embedded system. Often, control and communication paths within the system are tightly coupled to the actual physical configuration of the robot. Thus, programming a robot can hardly be achieved by experts of the domain in which the robot is...

This paper proposes a new concept to augment direct robot programming with sensor-based branching and looping. The objective is to derive a new robot programming paradigm in order to enable non-programmers to use robots for different tasks. Inspired by the ancient programming style based on the GOTO-statement, this approach requires the user to dea...

For effective collaboration between two humans, they are required to adapt to each other according to their apparent behavior in time. Similarly for intuitive human-robot collaboration the robot is required to adapt to the human intention apparent from his behavior. In this paper, we introduce a probability-based approach that helps the robot to ad...

Intention recognition plays a key role in the cooperation among the humans. An intention describes an action or sequence of actions to be performed for achieving the intended purpose. The cooperating humans learn each others intentions while cooperation. In this paper we propose three ways how a robot can learn the intention of the cooperating huma...

We introduce an approach of multi-view based surveillance of human/robot workspaces. Unknown objects within a robot workcell as e.g. humans are to be detected to avoid accidents with the robot. Therefore, two aspects are to be discussed: First, the reconstruction of regions within the robot workcell which are potentially occupied by unknown objects...

Fast manipulation of unknown objects is a basic ability on the way to the robotic co-worker, because it is mostly impractical to have a model of each object in the environment. This paper addresses the problem of grasping and delivering unknown objects by an industrial robot from a table using one single camera image. The delivery planning for each...

Humans have the capability of concept generalization. They can generalize an operation specific to an object on different objects present in the scene. In this paper we introduce a novel approach of rule based human intention generalization. The generalization is performed through Human-Robot Interaction (HRI) by inducing the rules online. The onli...

We present a human-robot-cooperation welding system, consisting of several phases including a coarse acquisition of the work piece geometry, interactive fine-scanning of welding seam regions and automated optimizing path planning after task specification by the user. The coarse geometry acquisition component is explained in detail showing the feasi...

In modern production facilities industrial robots and humans are supposed to interact sharing a common working area. In order to avoid collisions, the distances between objects need to be measured conservatively which can be done by a camera network. To estimate the acquired distance, unmodelled objects, e.g., an interacting human, need to be model...

We present a new approach that realizes an imagebasedfault tolerant distance computation for a multi-viewcamera system which conservatively approximates theshortest distance between unknown objects and 3Dvolumes. Our method addresses the industrial applicationof vision-based protective devices which are used to detectintrusions of humans into areas...

Combining the intelligent and situation dependent decision making capabilities of a human with the accuracy and power of a robot, performance of many tasks can be improved. The human-robot collaboration scenarios are increasing. Human-robot interaction is not only restricted to the humanoid robots interacting with the humans or to the mobile servic...

We present a general vision-based method for reconstructing multiple unknown objects (e.g. humans) within a known environment (e.g. tables, racks, robots) which usually has occlusions. These occlusions have to be explicitly considered since parts of the unknown objects might be hidden in some or even all camera views. In order to avoid cluttered re...

We introduce a general approach for surveillance of robots using depth images of multiple distributed smart cameras which can be either standard colour cameras or depth cameras. Unknown objects intruding the robot workcell are detected in the camera images and the minimum distance between the robot and all these detected unknown objects is calculat...

The purpose of this paper is to enable a developer to easily employ external sensors emitting a one-dimensional signal for flexible robot manipulation. To achieve this, the sensor must be calibrated using data tuples describing the relation between the positional change of the supervised object and the resulting sensor value. This information is us...

In current industrial applications without sensor surveillance, the robot workcell needs to be rather static. If the environment of the robot changes in an unplanned manner, e. g. a human enters the workcell and crosses the trajectory, a collision could result. Current research aims at relaxing the separation of robot and human workspaces. We prese...

The aim of this work is to show that robots can be improved using knowledge present or gained in previous executions of the task. Here, this knowledge is used to create search paths tailored to the problem at hand. We describe two approaches to create such paths automatically which enable robots to find an unknown position faster than standard path...

Dynamic environments, in which robots and for example humans share a common workspace impose a variety of requirements on path planning algorithms, including real-time capabilities and collision tests based on sensor input. We propose a randomized-roadmap-based path planning algorithm that limits the number of collision tests and distance calculati...

Programming by demonstration (PbD) is a promising approach to facilitate the use of robots. The bandwidth of PbD systems extends from simple play back to artificial intelligence methods. Our approach is based on the former. This paper aims to introduce and examine intuitive programming of spatial loops, i.e. program loops whose body changes accordi...

Today, real-time collision detection is a basic demand for many applications. While collision tests between known (modeled) objects have been around for quite a while, collision detection of known objects with dynamic, unknown (sensor-detected) objects remains a challenging field of research, especially when it comes to real-time requirements. The...

We present an anytime-capable fast deterministic greedy algorithm for smoothing piecewise linear paths consisting of connected linear segments. With this method, path points with only a small influence on path geometry (i.e. aligned or nearly aligned points) are successively removed. Due to the removal of less important path points, the computation...

The aim of this paper is to enable a programmer to easily employ external sensors for flexible robot manipulation. We describe a general approach to determine the relation between the position deviation of an object and the resulting data from sensors used to recognize this deviation. This information can be used to employ adaptation techniques to...

Surgical procedures with navigation or robot system support usually require pre-operative planning data. This data can be acquired with imaging techniques such as computed tomography (CT), the current gold standard due to its high precision. With such planning data, access trajectories, implant positions, individual milling paths etc. can be comput...

The aim of this paper is to enable a developer to easily employ external sensors for flexible robot manipulation in industrial applications. We outline a general approach to supervise robot tasks for workpiece drifts using external sensors. We describe two methods to detect workpiece drift independent from the type of sensor. Based on this, we expl...

Industrial robots are fast, powerful, enduring and precise. Humans on the other hand can handle complex manipulation task with superior quality and adapt flexibly to new and unknown situations. The combination of those strengths promises a huge potential for productivity increase in common industrial environments. The researchers of the project SIM...

We present a framework for an adaptive system capable of automatically optimizing a given sensor-based program for an industrial robot to facilitate the development of fast and robust applications. We present a systematic overview of the layers of the robot controller in which a speed-up may be gained by incorporating adaptation techniques. We then...

We present a path planner capable of efficient and real-time handling of known and unknown obstacles in highly dynamic workspaces. Known obstacles are acquired offline and stored in a world model, unknown obstacles are acquired online by one or multiple sensors. This is a typical situation for many applications. The method presented here exploits t...

We present a method for quickly determining the minimum distance between multiple known and multiple unkown objects within a camera image. Known objects are objects with known geometry, position, orientation, and configuration. Unkown objects are objects which have to be detected by a vision sensor but with unkown geometry, position, orientation an...

Currently industrial robotic systems employ almost exclusively global maps for navigation purposes, if any. Additional information - intra-process, spatial, current, and persistent sensor data - is useful to cope with uncertainty, measurement errors, and incompleteness of data. We propose to augment robot world models by using local sensors (which...

Abstract – Surgical procedures with navigation or robot system support usually require pre-operative plan- ning data. This data can be acquired with imaging tech- niques such as computed tomography (CT), the current gold standard due to its high precision. With such imaging data, access trajectories, implant positions, individual milling paths etc....