Abstract: SymAware addresses the fundamental need for a new conceptual framework for the quantification and mitigation of risks in a multi-agent system (MAS). The framework is expected to be compatible with the internal models and specifications of robotic agents and enables the safe simultaneous operation of collaborating autonomous agents and humans. The goal of SymAware is to provide theoretical tools and structural concepts for the agents to understand the uncertainty of the environment, be aware of the risks, and fulfill complex and dynamically changing tasks. The SymAware project will use compositional logic, symbolic computation, formal reasoning, and stochastic quantification to characterize and realize situational awareness of MAS in its various dimensions. Sustaining awareness by learning in social contexts, quantifying risks based on limited knowledge, and formulating risk-aware negotiation of task distributions will also be involved to enhance the risk-aware capability of a MAS with humans. The objectives of SymAware address the “Awareness Inside” challenge of the European Innovation Council by extending and formalizing human-based models of situational awareness and by providing a novel conceptual situational awareness framework for MASs that encompasses logical characterization and integrative formal reasoning of interdependent awareness dimensions including knowledge, spatiotemporal, risk and social dimensions. This will support transitioning to the safe mixed operation of autonomous agents and humans. [Details]
Period: 2022.10 - present
Role: Project leader as a postdoc
Principal Investigator: Dr. Ir. Sofie Haesaert
Affiliation: Eindhoven University of Technology (TU/e)
Funding Source: European Union Commission
Website: https://www.symaware.eu/
Consortium: Max Planck Institute (MPI), KTH Royal Institute of Technology (KTH), Uppsala University (UU), Siemens AG, Netherlands Aerospace Centre (NLR)
Abstract: AIARA is a cross-continent project that involves multiple partners from both Canada and Germany, including UBC and Kinova company from Canada, and DLR, ZAL, Fraunhofer, and Broetje from Germany. It receives funding from the Natural Sciences and Engineering Research Council of Canada (NSERC), the Consortium for Research and Innovation in Aerospace in Québec (CRIAQ), and the German Aeronautical Research Program LuFo (LuFo-8). The AIARA project seeks novel solutions for flexible and versatile robotic systems that can automatically adapt themselves to the changing conditions of the environment. Instead of the conventional approaches where a specific program is developed for a fixed robotic task in an unchanged environment, this project is intended to develop artificial-intelligence-enabled concepts and methods to enhance the adaptability of the robotic systems in variable environments and their extendability to new robotic tasks. Machine learning approaches, including but not limited to reinforcement learning, supervised learning, unsupervised learning, transfer learning, and meta-learning, will be used to construct adaptive models and develop adaptive methods for robotic manipulation tasks in variable environments. Hardware and software test benches will be developed to evaluate the feasibility and applicability of the approaches. Our ambition is to provide a novel direction toward reliable automation of the manufacturing and aerospace industry using highly adaptive robots enabled by cutting-edge artificial intelligence technologies. [Details]
Period: 2021.03 - 2023.09
Role: Project leader as a postdoc
Principal Investigator: Prof. Dr. Homayoun Najjaran
Affiliation: University of British Columbia (UBC), University of Victoria (Uvic)
Funding Source: Natural Sciences and Engineering Research Council of Canada (NSERC), Consortium for Research and Innovation in Aerospace in Quebec (CRIAQ)
Website: https://www.criaq.aero/en/projects/
Consortium: Kinova Company, German Aerospace Center (DLR), Fraunhofer-Gesellschaft, Center of Applied Aeronautical Research (ZAL), August Brötje GmbH
Abstract: The technical and technological advances that have been achieved over the past decades have led to a tremendous increase in both the types as well as the total amount of electrical and electronic equipment that is manufactured by the industry. On the other hand, the lowering of industrial production costs with the continuous and rapid change in technology has resulted in the widespread use of the produced devices in large quantities, along with the continuous need to often be upgraded or replaced. The above facts have led to the generation of enormous amounts of Waste Electrical and Electronic Equipment (WEEE). The importance of managing the WEEE materials and its tremendous impact on the economy, society, and environment is easy to be realized, by considering that the production of common electrical and electronic equipment (e.g. smartphones, PCs, monitors, tablets, home appliances, etc.) requires highly expensive and often rare material types (e.g. gold, copper, steel, etc.). Despite the great importance and the tremendous economic/environmental/societal impact of WEEE management, the current technology is still confronted with challenges, such as hazardous WEEE materials, human and environmental risks, illegal activities, recycling costs, and strict policies. Therefore, a ‘hybrid human-robot recycling plant for electrical and electronic equipment (HR-Recycler)’ operating in an indoor environment for the recycling of WEEE is desired. The ultimate goal of this project is to develop a sophisticated open human-robot working environment that will implement an HR-Recycler operating in an indoor setting. The fundamental principle behind the design of the envisaged system is the replacement of multiple currently manual, expensive, hazardous, and time-consuming tasks of WEEE materials pre-processing with correspondingly automatic robotic-based procedures, fused within a genuine human-robot collaboration context that will boost the productivity and quality of work in the plant. The primary output of the envisaged system will be to extract sorted electric/electronic device components and concentrated fractions of increased economic and environmental value; hence, contributing to the fundamental goal of the ‘European circular economy’ project and boosting economic activity in secondary markets. [Details]
Period: 2018.12 - 2020.11
Role: Participant as a doctoral candidate
Principal Investigator: PD Dr.-Ing. habil. Dirk Wollherr
Affiliation: Technical University of Munich
Funding Source: European Union Commission
Website: https://www.hr-recycler.eu/
Consortium: Centre for Research and Technology Hellas (CERTH), Institute for Bioengineering of Catalonia (IBEC), Tecnalia Research & Innovation (Tecnalia), Vrije Universiteit Brussel (VUB), COMAU, Diginext, GAIKER Technology Centre, Sadako Technologies, Robotnik, Baianat, Interecycling, Indumental.
Human-Robot Collaboration in Industrial Production
Abstract: Massive new changes in industrial production are expected to be witnessed as new requirements are proposed by Industry 4.0. These new requirements are expressed as the union of "lot-size one" (a product in this form is only produced in one piece, i.e. exactly once) and "mass customization" (each individual product in a large series is strongly adapted to the needs of a single user). The requirements for production automation have been changing in many industries for some time. The reason for this is the increasing variety of products to be manufactured and the ever-shorter product life cycles. In many cases, classic mass production in the true sense no longer takes place. In principle, the new requirements can be met in productions that essentially rely on human labor, mostly with lower quality than machines. Also and the strong production machinery adapted to the industrial products with relatively rigid processes is actually only for mass production. This poses a considerable challenge for conventional automation technology. In high-wage countries and in the face of demographic change, however, one is dependent on the increase in productivity that can be achieved through automation in order to be able to produce competitively. Based on the fact that the production processes that can be automated with conventional techniques are already automated, it is now necessary to promote automation to the next step to reach the required manual dexterity and intelligence that humans can achieve to date.
Such a strategy naturally leads to the concept of human-robot collaboration (HRC) which requires that the human labor and robots are not only coexisting in a common workspace but also accomplish the same industrial production task. This is based on a naive assumption that machines will not be able to completely replace humans in the foreseeable future, which is why seamless cooperation between humans and machines is of outstanding importance. Through cooperation, humans and robots can coordinate their actions and thus act as an effective team. Such an ability requires decision-making competence on the part of the robot, which enables a cooperative choice of action within a group of several humans and robots. The theoretical and conceptual foundation of HRC is built upon human-robot interaction (HRI) which has received much attention from the academic world in recent years, with promising partial concepts developed. However, a closed overall concept for the close cooperation between humans and robots in industrial production scenarios does not yet exist. The main challenges in this context arise from the information uncertainty, which can hardly be avoided in real scenarios, as well as the rapidly increasing problem complexity. This information uncertainty arises from the fact that the robot system cannot exactly know the exact state of the world, for example, the spatial position of task-relevant objects. Responsible for this is, on the one hand, residual errors in the evaluation of the sensors and, on the other hand, the fact that the activities of the human team members in particular cannot be fully observed. On the other hand, considerable previous knowledge is available, especially in the industrial assembly scenarios considered in this project. Both the properties and function of the objects relevant to the task and the actual assembly process in all details are known from the outset because they are recorded in the corresponding engineering systems as part of product development anyway. This previous knowledge should be used both when planning an assembly task and when monitoring the execution of the task by the human-robot team, for example, to interpret human actions and assign them to given elementary actions.
From this perspective, this project is actually aiming at providing a breaking new technical and scientific ground for the theoretical development and practical applications of HRC in industrial production. With this project, Siemens and the chair for control and regulation technology (LSR) at the Technical University of Munich are planning a cooperation to investigate the close human-robot collaboration in industrial production systems. [Details]
Period: 2015.10 - 2017.09
Role: Participant as a doctoral candidate
Principal Investigator: Prof. Dr.-Ing./Univ. Tokyo Martin Buss, PD Dr.-Ing. habil. Dirk Wollherr, Dr.-Ing. Georg von Wichert
Affiliation: Technical University of Munich (TUM)
Funding Source: Siemens AD-Campus project
Website: http://www.campus-ad.de/forschung/human-robot-collaboration-in-production/
Consortium: Technical University of Munich (TUM) and Siemens AG.