This study presents a procedure for the estimation of stopping behavior of industrial robots with a trained neural network. This trained network is presented as a single channel in a redundant architecture for safety control applications, where its potential for future integration with an analytical model of robot stopping is discussed. Basic physical relations for simplified articulated manipulators are derived, which motivate a choice of quantities to predict robot stopping behavior and inform the training and testing of a network for prediction of stopping distances and times. Robot stopping behavior is considered in the context of relevant standards ISO 10218-1, ISO/TS 15066 and IS0 13849-1, which inform the definitions for safety related stopping distances and times used in this study. Prior work on the estimation of robot stopping behavior is discussed alongside applications of machine learning to the broader field of industrial robotics, and particularly to the cases of prediction of forward and inverse kinematics with trained networks. A state-driven data collection program is developed to perform repeated stopping experiments for a controlled stop on path within a specified sampling domain. This program is used to collect data for a simulated and real robot system. Special attention is given to the identification of meaningful stopping times, which includes the separation of stopping into pre-deceleration and post-deceleration phases. A definition is provided for stopping of a robot in a safety context, based on the observation that residual motion over short distances (less than 1 mm) and at very low velocities (less than 1 mm/s) is not relevant to robot safety. A network architecture and hyperparameters are developed for the prediction of stopping distances and times for the first three joints of the manipulator without the inclusion of payloads. The result is a dual-network structure, where stopping distance predictions from the distance prediction network serve as inputs to the stopping time prediction network. The networks are validated on their capacity to interpolate and extrapolate predictions of robot stopping behavior in the presence of initial conditions not included in the training and testing data. A method is devised for the calculation of prediction errors for training training, testing and validation data. This method is applied both to interpolation and extrapolation to new initial velocity and positional conditions of the manipulator. In prediction of stopping distances and times, the network is highly successful at interpolation, resulting in comparable or nominally higher errors for the validation data set when compared to the errors for training and testing data. In extrapolation to new initial velocity and positional conditions, notably higher errors in the validation data predictions are observed for the networks considered. Future work in the areas of predictions of stopping behavior with payloads and tooling, further applications to collaborative robotics, analytical models of stopping behavior, inclusion of additional stopping functions, use of explainable AI methods and physics-informed networks are discussed. / Master of Science / As the uses for industrial robots continue to grow and expand, so do the need for robust safety measures to avoid, control, or limit the risks posed to human operators and collaborators. This is exemplified by Isaac Asimov's famous first law of robotics - "A robot may not injure a human being, or, through inaction, allow a human being to come to harm." As applications for industrial robots continue to expand, it is beneficial for robots and human operators to collaborate in work environments without fences. In order to ethically implement such increasingly complex and collaborative industrial robotic systems, the ability to limit robot motion with safety functions in a predictable and reliable way (as outlined by international standards) is paramount. In the event of either a technical failure (due to malfunction of sensors or mechanical hardware) or change in environmental conditions, it is important to be able to stop an industrial robot from any position in a safe and controlled manner. This requires real-time knowledge of the stopping distance and time for the manipulator. To understand stopping distances and times reliability, multiple independent methods can be used and compared to predict stopping behavior. The use of machine learning methods is of particular interest in this context due to their speed of processing and the potential for basis on real recorded data. In this study, we will attempt to evaluate the efficacy of machine learning algorithms to predict stopping behavior and assess their potential for implementation alongside analytical models. A reliable, multi-method approach for estimating stopping distances and times could also enable further methods for safety in collaborative robotics such as Speed and Separation Monitoring (SSM), which monitors both human and robot positions to ensure that a safe stop is always possible. A program for testing and recording the stopping distances and times for the robot is developed. As stopping behavior varies based on the positions and speeds of the robot at the time of stopping, a variety of these criteria are tested with the robot stopping program. This data is then used to train an artificial neural network, a machine learning method that mimics the structure of human and animal brains to learn relationships between data inputs and outputs. This network is used to predict both the stopping distance and time of the robot. The network is shown to produce reasonable predictions, especially for positions and speeds that are intermediate to those used to train the network. Future improvements are suggested and a method is suggested for use of stopping distance and time quantities in robot safety applications.

This thesis describes what the theatre is and what types of machinery can be found there. Then the issue of stage technical machinery is presented into the czech context, respectively into the Europian legislation with the emphasis on the technical norms usage, which are necessary to fulfil in order to reach this aim. In the next part of the thesis is made the analysis of control systems, which are used in the stage technics with the emphasis on the functionality of these control systems, since this functionality determines number and range of possible danger and dangerous events, later on. These dangerous situations are named and specified in detail, thereby their existence, importance and necessity of solution is being emphasized. There is also a possible solution outlined that these dangerous events can be solved in many cases by using of these programmable systems relating to the safety, thereby safety functions realized by these functions. Then the single steps are described into V-model with appropriate documents of these V-model steps. Outcome of this thesis will be suggested model of method and its development and software validation for programmable and control systems in theatre technology.

Several manufacturers of road vehicles today are working on developing autonomous vehicles. One subject that is often up for discussion when it comes to integrating autonomous road vehicles into the infrastructure is the safety aspect. There is in the context no common view of how safety should be quantified. As a contribution to this discussion we propose describing each potential hazardous event of a vehicle as a Semi-Markov Process (SMP). A reliability-based method for using the semi-Markov representation to calculate the probability of a hazardous event to occur is presented. The method simplifies the expression for the reliability using the Laplace-Stieltjes transform and calculates the transform of the reliability exactly. Numerical inversion algorithms are then applied to approximate the reliability up to a desired error tolerance. The method is validated using alternative techniques and is thereafter applied to a system for automated steering based on a real example from the industry. A desired evolution of the method is to involve a framework for how to represent each hazardous event as a SMP. / Flertalet tillverkare av vägfordon jobbar idag på att utveckla autonoma fordon. Ett ämne ofta på agendan i diskussionen om att integrera autonoma fordon på vägarna är säkerhet. Det finns i sammanhanget ingen klar bild över hur säkerhet ska kvantifieras. Som ett bidrag till denna diskussion föreslås här att beskriva varje potentiellt farlig situation av ett fordon som en Semi-Markov process (SMP). En metod presenteras för att via beräkning av funktionssäkerheten nyttja semi-Markov representationen för att beräkna sannolikheten för att en farlig situation ska uppstå. Metoden nyttjar Laplace-Stieltjes transformen för att förenkla uttrycket för funktionssäkerheten och beräknar transformen av funktionssäkerheten exakt. Numeriska algoritmer för den inversa transformen appliceras sedan för att beräkna funktionssäkerheten upp till en viss feltolerans. Metoden valideras genom alternativa tekniker och appliceras sedan på ett system för autonom styrning baserat på ett riktigt exempel från industrin. En fördelaktig utveckling av metoden som presenteras här skulle vara att involvera ett ramverk för hur varje potentiellt farlig situation ska representeras som en SMP.

Recent advancements in telecommunications have brought new tools about in the field of robotics, with offloading emerging as one of the most significant developments. Hence, computationally expensive tasks are performed on a server in the cloud instead of on the mobile robot, reducing processing costs in robots and enhancing their efficiency. However, one of the major challenges of offloading robot control is to maintain functional safety even when the connection with the server is interrupted. To mitigate these connectivity losses, an optimization-based method has been developed to compute an environment-dependent contingency plan. This plan is sent from the cloud to the robot together with the corresponding control command. The planner takes into account the current map, based on all sensor data collected up to the time of optimization, and the nominal trajectory to provide a sequence of safe control commands. Assuming that in the absence of connectivity, all detected objects will move at a constant speed. Therefore, the contingency plan would be executed on the robot only when connectivity to the cloud is lost, without making use of subsequent sensor data in the robot’s on-board processor. Thus, through the proposed method, it is possible to maximize the movement time of the mobile robot in case of loss of connectivity with the cloud controller without compromising any safety constraints. In this context, two different approaches have been designed based on the possibility of deviating from the nominal trajectory. In the first, called “path following”, the mobile robot is constrained to stay on the reference path, but can vary its speed, performing a safety brake when there is a risk of collision. In contrast, in “trajectory following”, deviation is allowed by trying to prolong the point at which the velocity is reduced. The evaluation shows that the optimal approach depends on the application for which the mobile robot will be used. Furthermore, these approaches do not overload the network bandwidth, since contingency plans can be optimized by parameterizing the velocity sequences or by reducing the sending rate through event-triggered sending. / De senaste framstegen inom telekommunikation har introducerat nya verktyg inom robotikens område, där offloading är en av de mest relevanta. Således utförs beäkningsintensiva uppgifter på en server i molnet istället för på den mobila roboten, vilket minskar bearbetningskostnaderna för roboter och ökar deras effektivitet. En av de största utmaningarna med att offloada robotstyrning är dock att bibehålla funktionell säkerhet även när anslutningen till fjärrservern bryts. För att hantera sådana avbrott, har vi utvecklat en optimeringsbaserad metod för att beräkna en reservplan, anpassad till miljön runt roboten. Denna plan skickas från molnet till roboten tillsammans med varje styrkommando. Planeraren beaktar den aktuella kartan, baserad på all sensordata som samlats in fram till nu, och den nominella banan och beräknar en säker reservplan i form av en sekvens av styrkommandon. För säkerhets skull antar planeraren att i händelse av ett avbrott, kommer alla hinder i kartan att närma sig roboten med en konstant hastighet. Det gör det säkert att exekvera reservplanen om anslutningen till molnet går förlorad, utan att använda efterföljande sensordata för att uppdatera kartan. Den föreslagna metoden gör det alltså möjligt att maximera tiden som den mobila roboten kan fortsätta köra vid förlust av anslutning till molnservern, utan att göra avkall på säkerheten. I detta projekt har vi utformat två olika planeringsmetoder, som skiljer sig vad gäller möjligheten att avvika från den nominella banan. I den första, kallad “path following”, tillåts inte roboten att avvika från referensbanan och utför därför en säkerhetsbromsning när det finns risk för kollision. I den andra, kallad “trajectory following”, tillåts roboten avvika från referensbanan, genom att försöka fördröja det ögonblick då roboten behöver bromsa. Utvärderingen visar att vilken metod som är bäst, beror på tillämpningen som den mobila roboten används för. Dessutom överbelastar dessa tillvägagångssätt inte nätverksbandbredden, eftersom beredskapsplaner kan optimeras genom att parameterisera hastighetssekvenser eller genom att minska överföringshastigheten. / Los recientes avances en las telecomunicaciones han traído consigo nuevas herramientas en la robótica, siendo el offloading una de los desarrollos más significativos. Así, las tareas computacionalmente más costosas se realizan en un servidor en la nube en lugar de en el robot móvil, reduciendo los costos de procesamiento en el robot y mejorando su eficiencia. Sin embargo, uno de los mayores desafíos del offloading de control de robots es mantener la seguridad funcional incluso cuando la conexión con el servidor se interrumpe. Con el fin de mitigar las pérdidas de conectividad, se ha desarrollado un método basado en optimizacion que calcula un plan de contingencia dependiente del entorno. Este plan se envía desde la nube al robot junto con el comando de control correspondiente. El planificador tiene en cuenta el mapa del entorno actual, basado en todos los datos del sensor recopilados hasta el momento de la optimización, y la trayectoria nominal para proporcionar una secuencia de comandos de control seguros. En este sentido, el planificador asume que, en ausencia de conectividad, todos los objetos detectados se aproximarán al robot a una velocidad constante. Este plan de contingencia se ejecutaría en el robot solo cuando se pierde la conectividad con la nube, sin hacer uso de datos de sensor posteriores en el procesador a bordo del robot. Por lo tanto, mediante el método propuesto, se logra maximizar el tiempo de movimiento del robot móvil en caso de pérdida de conectividad con el controlador en la nube sin sacrificar las restricciones de seguridad. En este contexto, dos enfoques distintos según la posibilidad de desviarse o no de la trayectoria nominal han sido diseñados. En el primero, denominado “path following”, no se permite que se desvíe de la referencia, aplicando un frenado de seguridad cuando existe riesgo de colisión. En cambio, en “trajectory following”, se permite la desviación para tratar de prolongar el momento en el que se reduce la velocidad. La evaluación muestra que el enfoque óptimo depende de la aplicación para la cual se utilizará el robot móvil. Además, estos enfoques no sobrecargan el ancho de banda de la red, ya que los planes de contingencia pueden optimizarse parametrizando las secuencias de velocidad o reduciendo la velocidad de envío.