A5003833130- Robot Manipulation and Learning
- Robotic Path Planning Algorithms
- Robotics and Sensor-Based Localization
- Social Robot Interaction and HRI
- Robotics and Automated Systems
- Manufacturing Process and Optimization
- AI-based Problem Solving and Planning
- Robotic Mechanisms and Dynamics
- Flexible and Reconfigurable Manufacturing Systems
- Action Observation and Synchronization
- Real-Time Systems Scheduling
- Autonomous Vehicle Technology and Safety
- Speech and dialogue systems
- 3D Surveying and Cultural Heritage
- Anomaly Detection Techniques and Applications
- Formal Methods in Verification
- Motor Control and Adaptation
- Digital Transformation in Industry
- 3D Shape Modeling and Analysis
- Semantic Web and Ontologies
- Real-time simulation and control systems
- Model-Driven Software Engineering Techniques
- Hand Gesture Recognition Systems
- Control and Dynamics of Mobile Robots
- Advanced Software Engineering Methodologies
University of Bamberg
2024
Technical University of Munich
2014-2023
Bayer (Germany)
2021
Fortiss
2013-2020
Institute of Electrical and Electronics Engineers
2019
Gorgias Press (United States)
2019
Vrije Universiteit Brussel
2019
Ludwig-Maximilians-Universität München
2008-2015
Information Technology University
2009
Embedded Systems (United States)
2008
Ethernet-based protocols are getting more and important for Industry 4.0 the Internet of Things. In this paper, we compare features, package overhead, performance some most in area. First, present a general feature comparison OPC UA, ROS, DDS, MQTT, followed by detailed wire protocol evaluation, which gives an overview over overhead establishing connection sending data. tests evaluate open-source implementations these measuring round trip time messages different system states: idle, high CPU...
In many future joint-action scenarios, humans and robots will have to interact physically in order successfully cooperate. Ideally, seamless human-robot interaction should not require training for the human, but be intuitively simple. Nonetheless, cooperation involve some degree of learning adaptation. Here, we report on a simple case physical interaction, hand-over task. Even such basic task as manually handing over an object from one agent another requires that both partners agree upon...
Current market demands require an increasingly agile production environment throughout many manufacturing branches. Traditional automation systems and industrial robots, on the other hand, are often too inflexible to provide economically viable business case for companies with rapidly changing products. The introduction of cognitive abilities into robotic is, therefore, a necessary step toward lean changeover seamless human-robot collaboration.
We present the exploring/exploiting tree (EET) algorithm for motion planning. The EET planner deliberately trades probabilistic completeness computational efficiency. This tradeoff enables to outperform state-of-the-art sampling-based planners by up three orders of magnitude. show that these considerable speedups apply a variety challenging real-world planning problems. performance improvements are achieved leveraging work space information continuously adjust sampling behavior planner. When...
Human activity recognition is crucial for intuitive cooperation between humans and robots. We present an approach applications in the context of human-robot interaction industrial settings. The based on spatial temporal features derived from skeletal data human workers performing assembly tasks. These were used to train a machine learning framework, which classifies discrete time frames with Random Forests subsequently models dependencies resulting states Hidden Markov Model. considered...
We present an approach for monitoring and interpreting human activities based on a novel multimodal vision-based interface, aiming at improving the efficiency of human-robot interaction (HRI) in industrial environments. Multi-modality is important concept this design, where we combine inputs from several state-of-the-art sensors to provide variety information, e.g. skeleton fingertip poses. Based typical workflows, derived multiple levels activity labels, including large-scale (e.g....
In this paper, we introduce an approach for leveraging CAD description to a semantic level, in order link additional knowledge models and exploit resulting synergy effects. This has been achieved by designing language, based on the Web Ontology Language (OWL), that is used define boundary representations (BREP) of objects. involves representing geometric entities meaningful way, e.g., circle defined coordinate frame radius instead set polygons. Furthermore, scope language also covers...
In this paper, we introduce a novel robot programming paradigm. It focuses on reducing the required expertise in robotics to level that allows shop floor workers use robots their application domain without need of extensive training. Our approach is user-centric and can interpret underspecified tasks, enabling communication an abstract level. Such high-level task descriptions make system amenable for users are experts particular domain, but have limited knowledge about thus not able specify...
This paper presents a concept of smart working environment designed to allow true joint-actions humans and industrial robots. The proposed system perceives its with multiple sensor modalities acts in it an robot manipulator assemble capital goods together human worker. In combination the reactive behavior robot, safe collaboration between is possible. Furthermore, anticipates behavior, based on knowledge databases decision processes, ensuring effective robot. As proof concept, we introduce...
Computationally efficient motion planning must avoid exhaustive exploration of configuration space. We argue that this can be accomplished most effectively by carefully balancing and exploitation. Exploration seeks to understand space, irrespective the problem, while exploitation acts solve problem given available information obtained exploration. present an exploring/exploiting tree (EET) planner balances its behavior. The acquires workspace subsequently uses for in If fails difficult...
Typical industrial workcells are composed of a plenitude devices from various manufacturers, which rely on their own specific control interfaces. To reduce setup and reconfiguration times, hardware-agnostic Plug & Produce system is required. In this paper, we present architecture that uses generic semantically augmented OPC UA skills for robots, tools, other components. Standardized skill interfaces parameters facilitate flexible component interchange automatic parametrization with focus...
Robot task planning is an inherently challenging problem, as it covers both continuous-space geometric reasoning about robot motion and perception, well purely symbolic knowledge actions objects. This paper presents a novel "knowledge of volumes" framework for solving generic tasks in partially known environments. In particular, this approach (abbreviated, KVP) combines the power symbolic, knowledge-level AI with efficient computation volumes, which serve intermediate representation action...
Today's industrial robots require expert knowledge and are not profitable for small medium sized enterprises with their lot sizes. It is our strong belief that more intuitive robot programming in an augmented reality work cell can dramatically simplify re-programming leverage robotics technology short production cycles. In this paper, we present a novel system defining virtual obstacles, specifying tool positions, tasks. We evaluate the user study and, specifically, investigate input of...
Due to the nonholonomic constraints of vehicle kinematics, parking and maneuvering a car in narrow clustered space are among most challenging driving tasks. This paper introduces an extended version Space Exploration Guided Heuristic Search (SEHS) method, called Orientation-Aware (OSEHS), solve path planning problems for maneuvering. method considers orientation exploration phase achieve knowledge about directions. Such information is exploited later heuristic search improve efficiency...
We propose a methodology for designing dependable Artificial Neural Networks (ANNs) by extending the concepts of understandability, correctness, and validity that are crucial ingredients in existing certification standards. apply concept concrete case study highway ANN-based motion predictor to guarantee safety properties such as impossibility ego vehicle suggest moving right lane if there exists another on its right.
Intuitive programming of industrial robots is especially important for small and medium-sized enterprises. We evaluated four different input modalities (touch, gesture, speech, 3D tracking device) regarding their preference, usability, intuitiveness robot programming.
The current trend to lot-size-one production requires reduced integration effort and easy reuse of available devices inside the line. These have offer a uniform interface fulfill these requirements.This paper presents hardware-agnostic skill model using semantic modeling capabilities OPC UA. provides standardized hardware or software functionality while offering an intuitive way grouping multiple skills higher hierarchical abstraction.Our is based on UA Programs modeled as open source...
The effectiveness of cognitive manufacturing systems in agile production environments heavily depend on the automatic assessment various levels interoperability between resources. For taking informed decisions, a semantically rich representation all resources workcell or line is required. OPC UA provides means for communication and information exchange such distributed settings.This paper proposes semantic resource's properties, which we use OWL ontologies to encode models that can be found...
This paper presents an approach to generate the time-optimal trajectory for a robot manipulator under certain kinematic constraints such as joint position, velocity, acceleration, and jerk limits. problem of generating that takes minimum time pass through specified waypoints is formulated nonlinear constraint optimization problem. Unlike prior approaches model motion consecutive Cubic Spline, we this with seven-segment acceleration profile, results in shorter overall while staying within...
Inferring a complete 3D geometry given an in-complete point cloud is essential in many vision and robotics applications. Previous work mainly relies on global feature extracted by Multi-layer Perceptron (MLP) for predicting the shape geometry. This suffers from loss of structural details, as its generator fails to capture detailed topology structure clouds using only features. The irregular nature makes this task more challenging. paper presents novel method completion address problem....
Intra-individual and inter-individual variations of key pinch strength were analysed in a standardised manner for healthy Caucasian adults (female n = 403; male 366) aged between 20 95 years. The mean was less women (right 6.6 kg; left 6.1 kg) than men 10.4 9.7 kg). Independently hand dominance or gender, the right side about 7% stronger. Constitutional variables such as forearm length, circumference size showed positive correlation with strength. Since age similar both genders, showing...