Planning is the sub-area of artificial intelligence that studies the process of selecting actions to lead an agent, e.g. a robot or a softbot, to a goal state. In many realistic scenarios, any choice of actions can lead the robot into a dead-end state, that is, a state from which the goal cannot be reached. In such cases, the robot can, pro-actively, resort to human help in order to reach the goal, an approach called symbiotic autonomy. In this work, we propose two different approaches to tackle this problem: (I) contingent planning, where the initial state is partially observable, configuring a belief state, and the outcomes of the robot actions are non-deterministic; and (II) probabilistic planning, where the initial state may be partially or totally observable and the actions have probabilistic outcomes. In both approaches, the human help is considered a scarce resource that should be used only when necessary. In contingent planning, the problem is to find a policy (a function mapping belief states into actions) that: (i) guarantees the agent will always reach the goal (strong policy); (ii) guarantees that the agent will eventually reach the goal (strong cyclic policy), or (iii) does not guarantee achieving the goal (weak policy). In this scenario, we propose a contingent planning system that considers human help to transform weak policies into strong (cyclic) policies. To do so, two types of human help are included: (i) human actions that modify states and/or belief states; and (ii) human observations that modify belief states. In probabilistic planning, the problem is to find a policy (a function mapping between world states and actions) that can be one of these two types: a proper policy, where the agent has probability 1 of reaching the goal; or an improper policy, in the case of unavoidable dead-ends. In general, the goal of the agent is to find a policy that minimizes the expected accumulated cost of the actions while maximizes the probability of reaching the goal. In this scenario, this work proposes probabilistic planners that consider human help to transform improper policies into proper policies however, considering two new (alternative) criteria: either to minimize the probability of using human actions or to minimize the expected number of human actions. Furthermore, we show that optimal policies under these criteria can be efficiently computed either by increasing human action costs or given a penalty when a human help is used. Solutions proposed in both scenarios, contingent planning and probabilistic planning with human help, were evaluated over a collection of planning problems with dead-ends. The results show that: (i) all generated policies (strong (cyclic) or proper) include human help only when necessary; and (ii) we were able to find policies for contingent planning problems with up to 10^15000 belief states and for probabilistic planning problems with more than 3*10^18 physical states. / Planejamento é a subárea de Inteligência Artificial que estuda o processo de selecionar ações que levam um agente, por exemplo um robô, de um estado inicial a um estado meta. Em muitos cenários realistas, qualquer escolha de ações pode levar o robô para um estado que é um beco-sem-saída, isto é, um estado a partir do qual a meta não pode ser alcançada. Nestes casos, o robô pode, pró-ativamente, pedir ajuda humana para alcançar a meta, uma abordagem chamada autonomia simbiótica. Neste trabalho, propomos duas abordagens diferentes para tratar este problema: (I) planejamento contingente, em que o estado inicial é parcialmente observável, configurando um estado de crença, e existe não-determinismo nos resultados das ações; e (II) planejamento probabilístico, em que o estado inicial é totalmente observável e as ações tem efeitos probabilísticos. Em ambas abordagens a ajuda humana é considerada um recurso escasso e deve ser usada somente quando estritamente necessária. No planejamento contingente, o problema é encontrar uma política (mapeamento entre estados de crença e ações) com: (i) garantia de alcançar a meta (política forte); (ii) garantia de eventualmente alcançar a meta (política forte-cíclica), ou (iii) sem garantia de alcançar a meta (política fraca). Neste cenário, uma das contribuições deste trabalho é propor sistemas de planejamento contingente que considerem ajuda humana para transformar políticas fracas em políticas fortes (cíclicas). Para isso, incluímos ajuda humana de dois tipos: (i) ações que modificam estados do mundo e/ou estados de crença; e (ii) observações que modificam estados de crenças. Em planejamento probabilístico, o problema é encontrar uma política (mapeamento entre estados do mundo e ações) que pode ser de dois tipos: política própria, na qual o agente tem probabilidade 1 de alcançar a meta; ou política imprópria, caso exista um beco-sem-saída inevitável. O objetivo do agente é, em geral, encontrar uma política que minimize o custo esperado acumulado das ações enquanto maximize a probabilidade de alcançar a meta. Neste cenário, este trabalho propõe sistemas de planejamento probabilístico que considerem ajuda humana para transformar políticas impróprias em políticas próprias, porém considerando dois novos critérios: minimizar a probabilidade de usar ações do humano e minimizar o número esperado de ações do humano. Mostramos ainda que políticas ótimas sob esses novos critérios podem ser computadas de maneira eficiente considerando que ações humanas possuem um custo alto ou penalizando o agente ao pedir ajuda humana. Soluções propostas em ambos cenários, planejamento contingente e planejamento probabilístico com ajuda humana, foram empiricamente avaliadas sobre um conjunto de problemas de planejamento com becos-sem-saida. Os resultados mostram que: (i) todas as políticas geradas (fortes (cíclicas) ou próprias) incluem ajuda humana somente quando necessária; e (ii) foram encontradas políticas para problemas de planejamento contingente com até 10^15000 estados de crença e para problemas de planejamento probabilístico com até 3*10^18 estados do mundo.

Que ce soit dans un contexte industriel ou quotidien, les robots deviennent de plus en plus présents dans notre environnement et sont désormais capables d'interagir avec des humains. Dans les milieux industriels, des robots viennent notamment assister les opérateurs des chaînes d'assemblage pour des tâches fatigantes et dangereuses. Robots et opérateurs sont alors amenés à partager le même espace physique (coprésence) et à effectuer des tâches en commun (collaboration). Alors que la sécurité des humains à proximité des robots doit être garantie à tout instant, il convient également de déterminer si le travail collaboratif est accepté par les opérateurs, en termes d'utilisabilité et d'utilité.La première problématique de la thèse consiste à déterminer quelles sont les composantes importantes rentrant en jeu dans l'acceptabilité de la collaboration homme-robot, du point de vue des opérateurs. Différents facteurs peuvent influencer cette acceptabilité : l'apparence des robots et leurs mouvements, la distance de sécurité ou encore le mode d'interaction avec le robot.Afin d'étudier le maximum de facteurs, nous proposons d'utiliser la réalité virtuelle pour mener des tests utilisateurs en environnement virtuel. Nous utilisons des questionnaires pour recueillir les impressions subjectives des opérateurs et des mesures physiologiques pour estimer leur état affectif (stress, effort). La deuxième problématique de la thèse consiste à déterminer si une méthodologie utilisant la réalité virtuelle est pertinente pour cette évaluation : les résultats issus des tests en environnement virtuel rendent-ils bien compte de la situation réelle ?Pour répondre aux problématiques de la thèse, trois cas d'étude ont été mis en place et quatre expérimentations ont été menées. Deux de ces expérimentations ont été reproduites à la fois en environnements réel et virtuel afin d'évaluer la pertinence des résultats issus de la situation virtuelle par rapport à la situation réelle. / Either in the context of the industry or of the everyday life, robots are becoming more and more present in our environment and are nowadays able to interact with humans. In industrial environments, robots now assist operators on the assembly lines for difficult and dangerous tasks. Then, robots and operators need to share the same physical space (copresence) and to manage common tasks (collaboration). On the one side, the safety of humans working near robots has to be guaranteed at all time. On the other hand, it is necessary to determine if such a collaborative work is accepted by the operators, in terms of usability and utility.The first problematic of the thesis consists in determining the important criteria that play a role in the acceptability, from the operators' point of view. Different factors can influence this acceptability: robot appearance, robot movements, safety distance or interaction modes with the robot.In order to study as many factors as possible, we intend to use virtual reality to perform user studies in virtual environments. We are using questionnaires to gather subjective impressions from operators and physiological measures to estimate their affective states (stress, effort). The second problematic of the thesis consists in determining if a methodology using virtual reality is relevant for this evaluation: can the results from studies in virtual environments be reproducible in equivalent physical situations?In order to answer the problematics of the thesis, three use cases have been implemented and four studies have been performed. Two of those studies rely on a physical situation and its virtual reality counterpart in order to evaluate the relevance of the results of the virtual situation compared to the real situation.

In Human-Robot Collaboration scenarios, safety must be ensured by a risk management process that requires the execution of computationally expensive perception models (e.g., based on computer vision) in real-time. However, robots usually have constrained hardware resources that hinder timely responses, resulting in unsafe operations. Although Multi-access Edge Computing allows robots to offload complex tasks to servers on the network edge to meet real-time requirements, this might not always be possible due to dynamic changes in the network that can cause congestion or failures. This work proposes a safety-based task offloading strategy to address this problem. The goal is to intelligently use edge resources to reduce delays in the risk management process and consequently enhance safety. More specifically, depending on safety and network metrics, a Reinforcement Learning (RL) solution is implemented to decide whether a less accurate model should run locally on the robot or a more complex one should run remotely on the network edge. A third possibility is to reuse the previous output through verification of temporal coherence. Experiments are performed in a simulated warehouse scenario where humans and robots have close interactions. Results show that the proposed RL solution outperforms the baselines in several aspects. First, the edge is used only when the network performance is good, reducing the number of failures (up to 47%). Second, the latency is also adapted to the safety requirements (risk X latency reduced up to 48%), avoiding unnecessary network congestion in safe situations and letting other robots in hazardous situations use the edge. Overall, the latency of the risk management process is largely reduced (up to 68%), and this positively affects safety (time in safe zone increased up to 3:1%). / I scenarier med människa-robotkollaboration måste säkerheten säkerställas via en riskhanteringsprocess. Denna process kräver exekvering av beräkningstunga uppfattningsmodeller (t.ex. datorseende) i realtid. Robotar har vanligtvis begränsade hårdvaruresurser vilket förhindrar att respons uppnås i tid, vilket resulterar i osäkra operationer. Även om Multi-access Edge Computing tillåter robotar att avlasta komplexa uppgifter till servrar på edge, för att möta realtidskraven, så är detta inte alltid möjligt på grund av dynamiska förändringar i nätverket som kan skapa överbelastning eller fel. Detta arbete föreslår en säkerhetsbaserad uppgiftsavlastningsstrategi för att hantera detta problem. Målet är att intelligent använda edge-resurser för att minska förseningar i riskhanteringsprocessen och följaktligen öka säkerheten. Mer specifikt, beroende på säkerhet och nätverksmätvärden, implementeras en Reinforcement Learning (RL) lösning för att avgöra om en modell med mindre noggrannhet ska köras lokalt eller om en mer komplex ska köras avlägset på edge. En tredje möjlighet är att återanvända sista utmatningen genom verifiering av tidsmässig koherens. Experimenten utförs i ett simulerat varuhusscenario där människor och robotar har nära interaktioner. Resultaten visar att den föreslagna RL-lösningen överträffar baslinjerna i flera aspekter. För det första används edge bara när nätverkets prestanda är bra, vilket reducerar antal fel (upp till 47%). För det andra anpassas latensen också till säkerhetskraven (risk X latens reducering upp till 48%), undviker onödig överbelastning i nätverket i säkra situationer och låter andra robotar i farliga situationer använda edge. I det stora hela reduceras latensen av riskhanterings processen kraftigt (upp till 68%) och påverkar på ett positivt sätt säkerheten (tiden i säkerhetszonen ökas upp till 4%).

The market has been evolving lately, with the introduction of more and more digital tools that industries are making use to improve their overall operations within the Production process. The integration of digital tools within the Product realization process has major advantages in improving production performance. Many large industries make use of digital tools to digitize their products making them smart products. Implementing these digital tools can be beneficial for reshaping the organization which can lead to better customer satisfaction and improve business strategy. This project explores different digital tools that can be integrated with the product realization process and how these tools contribute to the different production development phases. This thesis presents a detailed study of the digital tools Simulation, Visualization, Emulation, and Digital twins which can be integrated with the product development process. A pre-study is conducted to gather knowledge regarding the application of these tools and further discover how these tools can support the Product realization process and is used to describe which tool works best at which stage of the product realization process, which can be used to improve the efficiency and accuracy of the production process. Implementing these digital tools within the production facility can be associated to smart factory paradigm of the Fourth industrial revolution Industry4.0.This research aims to contribute to the use of digital tools in the production processes and aids in bridging the gap between traditional and modern manufacturing methods. The outcome of this study is to clarify how the above-mentioned digital tools are linked to the product realization process to support an efficient and digitalized production development, also mentioning the strengths and weaknesses of these tools. The resulting analysis has provided a framework developed to support an efficient digitalized production development and preparation process for assembly tasks utilizing human and robot collaboration. This research paper can be used as a guide for companies that want to explore how implementing digital tools in their product realization process and how it may improve their productivity.

Ursprünglich wurden Industrieroboter hauptsächlich hinter Schutzzäunen betrieben, um den Sicherheitsanforderungen gerecht zu werden. Mit der Flexibilisierung der Produktion wurden diese scharfen Trennbereiche zunehmend aufgeweicht und externe Sicherheitstechnik, wie Abstandssensoren, genutzt, um Industrieroboter schutzzaunlos zu betreiben. Ausgehend vom Gedanken dieser Koexistenz bzw. Kooperation wurde die Sicherheitssensorik in den Roboter integriert, um eine wirkliche Kollaboration zu ermöglichen. Diese sogenannten kollaborierenden Roboter, oder Cobots, eröffnen neue Applikationsfelder und füllen somit die bestehenden Automatisierungslücken. Doch welche Automatisierungsvariante ist aus wirtschaftlichen Gesichtspunkten die geeignetste? Bisherige Forschung untersucht zum Großteil isoliert eine der beiden Technologien, ohne dabei einen Systemvergleich hinsichtlich technologischer Spezifika und Wirtschaftlichkeit anzustellen. Daher widmet sich diese Dissertation einer Methodik zum wirtschaftlichen Vergleich von kollaborierenden Robotern und Industrierobotern in schutzzaunlosen Maschinenbeladungssystemen. Besonderer Fokus liegt dabei auf dem Herausarbeiten der technischen Faktoren, die die Wirtschaftlichkeit maßgeblich beeinflussen, um ein Systemverständnis der wirtschaftlichen Struktur beider Robotertechnologievarianten zu erhalten. Zur Untersuchung werden die Inhalte eines solchen Planungsvorhabens beschrieben, kategorisiert, systematisiert und modularisiert. Auf wirtschaftlicher Seite wird ein geeignetes Optimierungsmodell vorgestellt, während auf technischer Seite vor allem die Machbarkeit hinsichtlich Greifbarkeit, Layoutplanung, Robotergeschwindigkeiten und Zykluszeitbestimmung untersucht wird. Mit deduktiven, simulativen, empirischen und statistischen Methoden wird das Systemverhalten für die einzelnen Planungsinhalte analysiert, um die Gesamtwirtschaftlichkeit mit einem Minimum an Investment,- Produktions,- und Zykluszeitinformationen a priori vorhersagen zu können. Es wird gezeigt, dass durch einen Reverse Engineering Ansatz die notwendigen Planungsdaten, im Sinne von Layoutkomposition, Robotergeschwindigkeiten und Taktzeiten, mithilfe von Frontloading zu Planungsbeginn zur Verfügung gestellt werden können. Dabei dient der Kapitalwert als wirtschaftliche Bewertungsgrundlage, dessen Abhängigkeit vom Mensch-Roboter-Interaktionsgrad in einem Vorteilhaftigkeitsdiagramm für die einzelnen Technologiealternativen dargestellt werden kann. Wirtschaftlich fundierte Entscheidungen können somit auf quantitiativer Basis getroffen werden.:1. Introduction 25 1.1 Research Domain 25 1.2 Research Niche 26 1.3 Research Structure 28 2. State of the Art and Research 31 2.1 Turning Machines and Machine Tending 31 2.1.1 Tooling Machine Market Trends and Machine Tending Systems 31 2.1.2 Workpiece System 34 2.1.3 Machine System 36 2.1.4 Logistics System 39 2.1.5 Handling System 41 2.2 Robotics 43 2.2.1 Robot Installation Development and Application Fields 43 2.2.2 Fenceless Industrial and Collaborative Robots 48 2.2.3 Robot Grippers 55 2.3 Planning and Evaluation Methods 56 2.3.1 Planning of General and Manual Workstations 56 2.3.2 Cell Planning for Fully Automated and Hybrid Robot Systems 59 2.3.3 Robot Safety Planning 61 2.3.4 Economic Evaluation Methods 70 2.4 Synthesis - State of the Art and Research 71 3. Solution Approach 77 3.1 Need for Research and General Solution Approach 77 3.2 Use Case Delineation and Planning Focus 80 3.3 Economic Module – Solution Approach 86 3.4 Gripper Feasibility Module – Solution Approach 89 3.5 Rough Layout Discretization Model – Solution Approach 94 3.6 Cycle Time Estimation Module – Solution Approach 97 3.7 Collaborative Speed Estimation Module – Solution Approach 103 3.7.1 General Approach 103 3.7.2 Case 1: Quasi-static Contact with Hand 107 3.7.3 Case 2: Transient Contact with Hand 109 3.7.4 Case 3: Transient Contact with Shoulder 111 3.8 Synthesis – Solution Approach 114 4. Module Development 117 4.1 Economic Module – Module Development 117 4.1.1 General Approach 117 4.1.2 Calculation Scheme for Manual Operation 117 4.1.3 Calculation Scheme for Collaborative Robots 118 4.1.4 Calculation Scheme for Industrial Robots 120 4.2 Gripper Feasibility Module – Module Development 121 4.3 Rough Layout Discretization Module – Module Development 122 4.3.1 General Approach 122 4.3.2 Two-Dimensional Layout Pattern 123 4.3.3 Three-Dimensional Layout Pattern 125 4.4 Cycle Time Estimation Module – Module Development 126 4.4.1 General Approach 126 4.4.2 Reachability Study 127 4.4.3 Simulation Results 128 4.5 Collaborative Speed Estimation Module – Module Development 135 4.5.1 General Approach 135 4.5.2 Case 1: Quasi-static Contact with Hand 135 4.5.3 Case 2: Transient Contact with Hand 143 4.5.4 Case 3: Transient Contact with Shoulder 145 4.6 Synthesis – Module Development 149 5. Practical Verification 155 5.1 Use Case Overview 155 5.2 Gripper Feasibility 155 5.3 Layout Discretization 156 5.4 Collaborative Speed Estimation 157 5.5 Cycle Time Estimation 158 5.6 Economic Evaluation 160 5.7 Synthesis – Practical Verification 161 6. Results and Conclusions 165 6.1 Scientific Findings and Results 165 6.2 Critical Appraisal and Outlook 173 / Initially, industrial robots were mainly operated behind safety fences to account for the safety requirements. With production flexibilization, these sharp separation areas have been increasingly softened by utilizing external safety devices, such as distance sensors, to operate industrial robots fenceless. Based on this idea of coexistence or cooperation, safety technology has been integrated into the robot to enable true collaboration. These collaborative robots, or cobots, open up new application fields and fill the existing automation gap. But which automation variant is most suitable from an economic perspective? Present research dealt primarily isolated with one technology without comparing these systems regarding technological and economic specifics. Therefore, this doctoral thesis pursues a methodology to economically compare collaborative and industrial robots in fenceless machine tending systems. A particular focus lies on distilling the technical factors that mainly influence the profitability to receive a system understanding of the economic structure of both robot technology variants. For examination, the contents of such a planning scheme are described, categorized, systematized, and modularized. A suitable optimization model is presented on the economic side, while the feasibility regarding gripping, layout planning, robot velocities, and cycle time determination is assessed on the technical side. With deductive, simulative, empirical, and statistical methods, the system behavior of the single planning entities is analyzed to predict the overall profitability a priori with a minimum of investment,- production,- and cycle time information. It is demonstrated that the necessary planning data, in terms of layout composition, robot velocities, and cycle times, can be frontloaded to the project’s beginning with a reverse engineering approach. The net present value serves as the target figure, whose dependency on the human-robot interaction grade can be illustrated in an advantageousness diagram for the individual technical alternatives. Consequently, sound economic decisions can be made on a quantitative basis.:1. Introduction 25 1.1 Research Domain 25 1.2 Research Niche 26 1.3 Research Structure 28 2. State of the Art and Research 31 2.1 Turning Machines and Machine Tending 31 2.1.1 Tooling Machine Market Trends and Machine Tending Systems 31 2.1.2 Workpiece System 34 2.1.3 Machine System 36 2.1.4 Logistics System 39 2.1.5 Handling System 41 2.2 Robotics 43 2.2.1 Robot Installation Development and Application Fields 43 2.2.2 Fenceless Industrial and Collaborative Robots 48 2.2.3 Robot Grippers 55 2.3 Planning and Evaluation Methods 56 2.3.1 Planning of General and Manual Workstations 56 2.3.2 Cell Planning for Fully Automated and Hybrid Robot Systems 59 2.3.3 Robot Safety Planning 61 2.3.4 Economic Evaluation Methods 70 2.4 Synthesis - State of the Art and Research 71 3. Solution Approach 77 3.1 Need for Research and General Solution Approach 77 3.2 Use Case Delineation and Planning Focus 80 3.3 Economic Module – Solution Approach 86 3.4 Gripper Feasibility Module – Solution Approach 89 3.5 Rough Layout Discretization Model – Solution Approach 94 3.6 Cycle Time Estimation Module – Solution Approach 97 3.7 Collaborative Speed Estimation Module – Solution Approach 103 3.7.1 General Approach 103 3.7.2 Case 1: Quasi-static Contact with Hand 107 3.7.3 Case 2: Transient Contact with Hand 109 3.7.4 Case 3: Transient Contact with Shoulder 111 3.8 Synthesis – Solution Approach 114 4. Module Development 117 4.1 Economic Module – Module Development 117 4.1.1 General Approach 117 4.1.2 Calculation Scheme for Manual Operation 117 4.1.3 Calculation Scheme for Collaborative Robots 118 4.1.4 Calculation Scheme for Industrial Robots 120 4.2 Gripper Feasibility Module – Module Development 121 4.3 Rough Layout Discretization Module – Module Development 122 4.3.1 General Approach 122 4.3.2 Two-Dimensional Layout Pattern 123 4.3.3 Three-Dimensional Layout Pattern 125 4.4 Cycle Time Estimation Module – Module Development 126 4.4.1 General Approach 126 4.4.2 Reachability Study 127 4.4.3 Simulation Results 128 4.5 Collaborative Speed Estimation Module – Module Development 135 4.5.1 General Approach 135 4.5.2 Case 1: Quasi-static Contact with Hand 135 4.5.3 Case 2: Transient Contact with Hand 143 4.5.4 Case 3: Transient Contact with Shoulder 145 4.6 Synthesis – Module Development 149 5. Practical Verification 155 5.1 Use Case Overview 155 5.2 Gripper Feasibility 155 5.3 Layout Discretization 156 5.4 Collaborative Speed Estimation 157 5.5 Cycle Time Estimation 158 5.6 Economic Evaluation 160 5.7 Synthesis – Practical Verification 161 6. Results and Conclusions 165 6.1 Scientific Findings and Results 165 6.2 Critical Appraisal and Outlook 173

In human-robot collaborative (HRC) scenarios where humans and robots work together sharing the same workspace, there is a risk of potential hazard that may occur. In this work, an AI-based risk analysis solution has been developed to identify any condition that may harm a robot and its environment. The information from the risk analysis is used in a risk mitigation module to reduce the possibility of being in a hazardous situation. The goal is to develop safety for HRC scenarios using different AI algorithms and to check the possibilities of improving efficiency of the system without any compromise on the safety. This report presents risk mitigation strategies that were built on top of the robot’s control system and based on the ISO 15066 standard. Each of them used semantic information (scene graph) about the robot’s environment and changed the robot’s movement by scaling speed. The first implementation of risk mitigation strategy used Fuzzy Logic System. This system analyzed the riskiest object’s properties to adjust the speed of the robot accordingly. The second implementation used Reinforcement Learning and considered every object’s properties. Three networks (fully connected network, convolutional neural network, and hybrid network) were implemented to estimate the Qvalue function. Additionally, local and edge computation architecture wereimplemented to measure the computational performance on the real robot. Each model was evaluated by measuring the safety aspect and the performance of the robot in a simulated warehouse scenario. All risk mitigation modules were able to reduce the risk of potential hazard. The fuzzy logic system was able to increase the safety aspect with the least efficiency reduction. The reinforcement learning model had safer operation but showed a more compromised efficiency than the fuzzy logic system. Generally, the fuzzy logic system performed up to 28% faster than reinforcement learning but compromised up to 23% in terms of safety (mean risk speed value). In terms of computational performance, edge computation was performed faster than local computation. The bottleneck of the process was the scene graph generation which analyzed an image to produce information for safety analysis. It took approximately 15 seconds to run the scene graph generation on the robot’s CPU and 0.3 seconds on an edge device. The risk mitigation module can be selected depending on KPIs of the warehouse operation while the edge architecture must be implemented to achieve a realistic performance. / I HRC-scenarier mellan människor och robotar där människor och robotar arbetar tillsammans och delar samma arbetsyta finns det risk för potentiell fara som kan uppstå. I detta arbete har en AI-baserad lösning för riskanalys utvecklats för att identifiera alla tillstånd som kan skada en robot och dess miljö. Informationen från riskanalys används i en riskreduceringsmodul för att minska risken för att vara i en farlig situation. Målet är att utveckla säkerhet för HRC-scenarier med olika AI-algoritmer och att kontrollera möjligheterna att förbättra systemets effektivitet utan att kompromissa med säkerheten.Denna rapport presenterar strategier för riskreducering som byggdes ovanpå robotens styrsystem och baserade på ISO 15066-standarden. Var och en av dem använder semantisk information (scendiagram) om robotens miljö och förändrar robotens rörelse genom skalning av hastighet. Den första implementetationen av riskreducerande strategi använder Fuzzy Logic System. Detta system analyserade de mest riskabla objektens egenskaper för att justera robotens hastighet i enlighet därmed. Den andra implementeringen använder förstärkningslärande och betraktade varje objekts egenskaper. Tre nätverk (fully connected network, convolutional neural network, and hybrid network) implementeras för att uppskatta Q-värde-funktionen. Dessutom implementerade vi också lokaloch edge-arkitektur för att beräkna beräkningsprestanda på den verkliga roboten. Varje modell utvärderas genom att mäta säkerhetsaspekten och robotens prestanda i ett simulerat lagerscenario. Alla riskreduceringsmoduler kunde minska risken för potentiell fara. Fuzzy logicsystem kunde öka säkerhetsaspekten med minsta effektivitetsminskning. Förstärkningsinlärningsmodellen har säkrare drift men har en mer begränsad effektivitet än det fuzzy logiska systemet. I allmänhet fungerar fuzzy logicsystem upp till 28 % snabbare än förstärkningslärande men komprometterar upp till 23 % när det gäller säkerhet (medelrisk hastighetsvärde). När det gäller beräkningsprestanda utfördes kantberäkningen snabbare än lokal beräkning. Flaskhalsen för processen var scengrafgenerering som analyserade en bild för att producera information för säkerhetsanalys. Det tog cirka 15 sekunder att köra scengrafgenerering på robotens CPU och 0,3 sekunder på en kantenhet. Modulen för riskreducering kan väljas beroende på KPI för lagerdriften medan edge-arkitekturen måste implementeras för att uppnå en realistisk prestanda.

Purpose: The thesis investigates the automation of screwing operation on moving assembly line, that can be integrated with the current workflow using automation technology and collaborative robots. The study focuses on reducing the number of manual tasks in screwing operation while considering the ergonomics and safety factor of the operator.    Method: The study uses mixed methods research approach, such as interviews and observation, along with literature review on the desired topics. The study focus on details of the company and their production layout. Practical challenges are derived from the combination of data, providing a guidance on developing an automated solution.    Findings: The research questions were answered, using the data derived from the interviews. The empirical data was compared with the theoretical data to enhance the validity. The resulted data identified to develop a new concept to satisfy all the technical functions.    Implications: The research finds out that a collaboration with an industrial robot, can help the operators to perform desired task. The robot needs to be supported with other aspects such as motion of robot, monitor system and safety. The research identified a conceptual design to overcome these challenges, and to change the production layout to accommodate them is recommended.    Delimitations: The study is focused on the screwing technology on a moving assembly line, and a concept is developed to overcome the challenges. There is no prototype constructed, the proposed idea is designed in robot studio to showcase the idea.    Keyword: Moving assembly line, Human-robot collaboration, Screwing technology, Automated screwing technology, industrial robots, Automobile manufacturing

BACKGROUND: The integration of robotics, and mixed reality has ushered in a substantial revolution within the realm of Industry 4.0. The incorporation of robots into the manufacturing sector plays a pivotal role in enhancing productivity, in which humans and robots collaborate with each other. However, the current robotic system operates within predefined pathways exclusively, lacking an automated mechanism for identifying obstacle free routes to facilitate the movement of robot . Also, in the Human Robot Collaboration , there exists a deficiency in visualising robot motion and status, consequently arise safety vulnerabilities for human operators. OBJECTIVES: This thesis aims to implement a pathfinding algorithm for the robot movement using a mixed reality environment. This Mixed Reality application is used to assign targets and handle obstacles in the robot movement path. The visual guide about the robot movement path, the state of the robot and the tasks to the user that will be displayed using MR. METHODS : In pursuit of the thesis objectives, a Mixed Reality environment was developed using Unity alongside MRTK plugins. Within this framework, an A Star pathfinding algorithm was implemented, facilitating the computation of obstacle free routes between source and destination points. This MR environment not only visualises the trajectory of the robot 's movement but also presents robot status updates and an intuitive interface for operator robot communication. The development process involved creating essential code using C# within the Visual Studio IDE. This code was subsequently deployed onto the HoloLens 2, the designated hardware device for MR applications. The positioning and alignment of virtual objects in relation to the physical world were achieved using the QR code methodology. In this context, source and destination points for the robot 's movement were symbolised as targets, while obstacles were represented by square game objects. For the control and communication of the ABB GoFa C RB 15000 robot, RAPID code was devised within Robot Studio.To guide the thesis, a constructivist philosophical paradigm was embraced, aiming to enhance efficacy. Ethical considerations were scrupulously considered for data collection, prioritising user privacy within the MR environment. Furthermore, commitment to sustainability was maintained throughout the thesis work, yielding environmental, economic, and societal advantages. ANALYSIS: The project that was developed underwent analysis through the scenarios, including both obstacle laden and obstacle free pathfinding situations. The A Star pathfinding algorithm, effectively calculated the obstacle free routes between targets and accomplished designated robotic tasks. This implementation not only offered visual path guidance but also supplied status updates. The analysis process involved observations, video recordings, and documentation. The findings indicated that the created Mixed Reality environment indeed enhanced safety and cognitive ergonomics for the operator. This section also outlines the industrial applications of the project developed. CONCLUSION: Successful development of a Mixed Reality environment has been achieved, aimed at enabling automated obstacle free pathfinding. This environment also offers visualisations for path and status information, with the goal of enhancing safety and cognitive ergonomics in Human Robot Collaboration. Throughout this thesis endeavour, strong attention has been paid to ethical considerations and sustainability.

Nowadays, the use of robots in industrial tasks is growing constantly. However, manual assembly is one area that is hard to make fully automated since manual assembly operations work with different shapes and products that require human finesse to do some operations. Humans, on the other hand, have a lot of limitations since this kind of task can be unergonomic and repetitive for operators, which can cause them stress, fatigue, repetitive stress injuries(RSI), and repetitive motion injuries. This project involved designing an insert for the gripper 2F-85 (version 3) that would allow the collaborative robot (UR5) to carry out more assembly tasks in order to relieve human workers of repetitive tasks. The insert has to handle cylindrical shapes in addition to bigger parts that the actual insert cannot handle due to its parallel stroke. For that, a detailed market analysis and insert research were conducted in the initial study. The new insert was then developed using a double-diamond design process. The needs were ranked using the Moscow prioritization method, and ideas were then generated using the brainstorming technique. The final concept was chosen using the weighted decision matrix method. After the final concept selection, computer-aided design (CAD) technology was employed to create the new insert's 3D model and its technical specifications. The mechanical behaviour of the new insert was analysed to reflect its range of workability, expressing the maximum force that it can withstand on each of its grip work surfaces without presenting plastic deformation. For this study, finite element analyses were conducted following the general method for linear structural analysis using Abaqus. Achieving an insert that can reach, transport, and assemble different shapes will help integrate collaborative robots into manual assembly processes, avoiding the cost of a new gripper.

Human-Robot Collaboration (HRC), where both humans and robots work in the same environment simultaneously, is an emerging field and has increased massively during the past decade. For this collaboration to be feasible and safe, robots need to perform a proper safety analysis to avoid hazardous situations. This safety analysis procedure involves complex computer vision tasks that require a lot of processing power. Therefore, robots with constrained computing resources cannot execute these tasks without any delays, thus for executing these tasks they rely on edge infrastructures, such as remote computational resources accessible over wireless communication. In some cases though, the edge may be unavailable, or connection to it may not be possible. In such cases, robots still have to navigate themselves around the environment, while maintaining high levels of safety. This thesis project focuses on reducing the complexity and the total number of parameters of pre-trained computer vision models by using model compression techniques, such as pruning and knowledge distillation. These model compression techniques have strong theoretical and practical foundations, but work on their combination is limited, therefore it is investigated in this work. The results of this thesis project show that in the test cases, up to 90% of the total number of parameters of a computer vision model can be removed without any considerable reduction in the model’s accuracy. / Människa och robot samarbete (förkortat HRC från engelskans Human-Robot Collaboration), där både människor och robotar arbetar samtidigt i samma miljö, är ett växande forskningsområde och har ökat dramatiskt över de senaste decenniet. För att detta samarbetet ska vara möjligt och säkert behöver robotarna genomgå en ordentlig säkerhetsanalys så att farliga situationer kan undvikas. Denna säkerhetsanalys inkluderar komplexa Computer Vision uppgifter som kräver mycket processorkraft. Därför kan inte robotar med begränsad processorkraft utföra dessa beräkningar utan fördröjning, utan måste istället förlita sig på utomstående infrastruktur för att exekvera dem. Vid vissa tillfällen kan dock denna utomstående infrastruktur inte finnas på plats eller vara svår att koppla upp sig till. Även vid dessa tillfällen måste robotar fortfarande kunna navigera sig själva genom en lokal, och samtidigt upprätthålla hög grad av säkerhet. Detta projekt fokuserar på att reducera komplexiteten och det totala antalet parametrar av för-tränade Computer Vision-modeller genom att använda modellkompressionstekniker så som: Beskärning och kunskapsdestilering. Dessa modellkompressionstekniker har starka teoretiska grunder och praktiska belägg, men mängden arbeten kring deras kombinerade effekt är begränsad, därför är just det undersökt i detta arbetet. Resultaten av det här projektet visar att up till 90% av det totala antalet parametrar hos en Computer Vision-modell kan tas bort utan någon noterbar försämring av modellens säkerhet.