This thesis investigates the use of monocular computer vision for odometry, specificallyemploying optical flow techniques. The goal is to develop and evaluate avisual odometry system that accurately estimates the trajectory and rotation of acamera in real-time. The system utilizes a camera and the Lucas-Kanade methodto capture high-frame-rate images, detect and track features within these images,and calculate the camera’s motion based on the movement of these features.Odometry is essential for estimating the position and orientation of moving objects,such as robots or vehicles, over time. Traditional methods rely on wheel encodersand Inertial Measurement Units (IMU), but visual odometry leverages visual datato enhance accuracy and robustness, without the risk of slippage and change inwheel diameter from loads. Furthermore a visual odometry system, like the oneused in this project, is not affected by occlusion.In this project, a camera was set up and calibrated, followed by the implementationof feature detection using the Shi-Tomasi corner detection algorithm. TheLucas-Kanade method was then applied to estimate optical flow, and an affinetransformation was used to compute the translation and rotation of the camera.The system’s performance was evaluated based on accuracy, computational efficiency,and robustness to noise.The results demonstrate that the visual odometry system can effectively track thecamera’s motion with a high degree of accuracy. But with limitations in speed. Thediscussion highlights potential applications in autonomous navigation and areas forfuture improvement, such as integrating additional sensors and enhancing featuredetection algorithm. / Denna rapport undersöker användningen av monokulärt datorseende förodometri, med särskilt fokus på optisk flödesteknik. Målet är att utveckla ochutvärdera ett visuellt odometrisystem som noggrant uppskattar en kameras positionoch rotation i realtid. Systemet använder en kamera och Lucas-Kanademetodenför att fånga bilder i hög frekvens, upptäcka och spåra pixelpunkter inomdessa bilder samt beräkna kamerans rörelse baserat på dessa punkters förflyttning.Odometri är viktigt för att uppskatta position och orientering av rörliga objekt,såsom robotar eller fordon, över tid. Traditionella metoder förlitar sig på hjulenkodersoch Inetrial Measurement Units (IMU), men visuell odometri utnyttjarvisuella data för att förbättra noggrannheten och robustheten, utan risken förslirning och förändringar i hjuldiametern på grund av belastningar. Dessutompåverkas inte ett visuellt odometrisystem, som det som används i detta projekt,av ocklusion.I detta projekt installerades och kalibrerades en kamera, följt av implementeringav funktionsdetektion med hjälp av Shi-Tomasi hörndetektionsalgoritm.Därefter tillämpades Lucas-Kanade-metoden för att uppskatta optiskt flöde, ochen affin transformation användes för att beräkna kamerans förflyttning och rotation.Systemets prestanda utvärderades utifrån noggrannhet, beräkningseffektivitetoch robusthet mot brus.Resultaten visar att det visuella odometrisystemet effektivt kan spåra kameransrörelse med hög noggrannhet. Med begränsning beroende på hastighet. Diskussionenbelyser potentiella tillämpningar inom autonom navigering och områden förframtida förbättringar, såsom integrering av ytterligare sensorer och förbättringav feature detection algoritmen.

The pursuit of multi-agent exploration is driven by its capacity to enhance operational robustness and efficiency in complex, dynamic environments, paving the way for advancements in autonomous systems and robotics. This thesis explores the development and assessment of decentralised planning algorithms within multi-agent systems, using the CARLA simulation environment. A methodology combining simulation-based testing and theoretical analysis was employed to evaluate the efficiency, and scalability of various decentralised planning strategies.  The study systematically analysed three different exploration strategies for multi-agent systems: Greedy, MinPos, and Hungarian Assignments, across various configurations concerning the number of agents and communication demands. The Hungarian Assignment strategy demonstrates the highest efficiency in area coverage and coordination, especially as the number of agents increases. Meanwhile, the Greedy Assignment strategy requires the least communication bandwidth, indicating its potential for scenarios with limited communication capabilities. The MinPos Assignment, while facilitating better spatial distribution of agents than the Greedy Assignment, showed a moderate increase in communication demands and did not significantly outperform the Greedy Assignment in terms of efficiency. This work contributes to the field by providing insights into the trade-offs between exploration efficiency and communication overhead in multi-agent systems. Future work could explore synchronisation mechanisms, collision-avoidance strategies, and further decentralisation of the system's components.

This thesis and appended papers present the process of tacking the problem of environment modeling for autonomous agent. More specifically, the focus of the work has been semantic mapping of warehouses. A semantic map for such purpose is expected to be layout-like and support semantics of both open spaces and infrastructure of the environment. The representation of the semantic map is required to be understandable by all involved agents (humans, AGVs and WMS.) And the process of semantic mapping is desired to lean toward full-autonomy, with minimum input requirement from human user. To that end, we studied the problem of semantic annotation over two kinds of spatial map from different modalities. We identified properties, structure, and challenges of the problem. And we have developed representations and accompanied methods, while meeting the set criteria. The overall objective of the work is “to develop and construct a layer of abstraction (models and/or decomposition) for structuring and facilitate access to salient information in the sensory data. This layer of abstraction connects high level concepts to low-level sensory pattern.” Relying on modeling and decomposition of sensory data, we present our work on abstract representation for two modalities (laser scanner and camera) in three appended papers. Feasibility and the performance of the proposed methods are evaluated over data from real warehouse. The thesis conclude with summarizing the presented technical details, and drawing the outline for future work. / Automatic Inventory and Mapping of Stock (AIMS)

The world contains objects with various properties - rigid, granular, liquid, elastic or plastic. As humans, while interacting with the objects, we plan our manipulation by considering their properties. For instance, while holding a rigid object such as a brick, we adapt our grasp based on its centre of mass not to drop it. On the other hand while manipulating a deformable object, we may consider additional properties to the centre of mass such elasticity, brittleness etc. for grasp stability. Therefore, knowing object properties is an integral part of skilled manipulation of objects.  For manipulating objects skillfully, robots should be able to predict the object properties as humans do. To predict the properties, interactions with objects are essential. These interactions give rise distinct sensory signals that contains information about the object properties. The signals coming from a single sensory modality may give ambiguous information or noisy measurements. Hence, by integrating multi-sensory modalities (vision, touch, audio or proprioceptive), a manipulated object can be observed from different aspects and this can decrease the uncertainty in the observed properties. By analyzing the perceived sensory signals, a robot reasons about the object properties and adjusts its manipulation based on this information. During this adjustment, the robot can make use of a simulation model to predict the object behavior to plan the next action. For instance, if an object is assumed to be rigid before interaction and exhibit deformable behavior after interaction, an internal simulation model can be used to predict the load force exerted on the object, so that appropriate manipulation can be planned in the next action. Thus, learning about object properties can be defined as an active procedure. The robot explores the object properties actively and purposefully by interacting with the object, and adjusting its manipulation based on the sensory information and predicted object behavior through an internal simulation model. This thesis investigates the necessary mechanisms that we mentioned above to learn object properties: (i) multi-sensory information, (ii) simulation and (iii) active exploration. In particular, we investigate these three mechanisms that represent different and complementary ways of extracting a certain object property, the deformability of objects. Firstly, we investigate the feasibility of using visual and/or tactile data to classify the content of a container based on the deformation observed when a robotic hand squeezes and deforms the container. According to our result, both visual and tactile sensory data individually give high accuracy rates while classifying the content type based on the deformation. Next, we investigate the usage of a simulation model to estimate the object deformability that is revealed through a manipulation. The proposed method identify accurately the deformability of the test objects in synthetic and real-world data. Finally, we investigate the integration of the deformation simulation in a robotic active perception framework to extract the heterogenous deformability properties of an environment through physical interactions. In the experiments that we apply on real-world objects, we illustrate that the active perception framework can map the heterogeneous deformability properties of a surface. / <p>QC 20170517</p>

The GE Healthcare Life sciences organization develop and manufacture bioreactors, mixers, filtration skids and chromatography systems used together in a biomanufacturing platform. The platform is monitored and controlled by a distributed control system through a Programmable Logic Controller (PLC). The automation software controlling the platform is today tested and verified together with the physical units. The software use PROFIBUS, an industry standard for industrial automation, for communication and control of the units. Limited access to the physical units is usually a bottleneck and it's difficult to test abnormal situations to make sure the correct alarms are triggered. To reduce the hardware dependency and to provide support during test and verification, a virtual environment is developed to simulate the behavior of a bioreactor during execution. A .NET application has been developed together with a mathematical framework to simulate a cell culture and to return relevant process parameters such as pH, dissolved oxygen, temperature and weight. The results show that it's possible to simulate a bioreactor and to communicate with the control system. The software can be a valuable tool when developing and testing automation software but should not be used for process optimization or tuning of control parameters.

A controlled riderless bicycle is desired for the purpose of testing autonomous vehicles ability to detect and recognise cyclists. The bicycle, which is a highly unstable system with complex dynamics have been subject to research for over a century, and in the last decades, controllers have been developed for autonomous bicycles. The controllers are often only evaluated in simulation, but some complex controllers have been developed on real-life bicycles as well. The goal of this work is to validate sensors and subsystems of an instrumented bicycle and to develop a robust controller which can balance a bicycle by using actuation on the steering axis alone. Using an iterative design process, the sensor measuring the lean angle and the steering system are improved and validated. By sensing the lean angle, the handlebar is manipulated to make the bicycle stable. For this purpose, a P, PD, two different PID, an LQR and a fuzzy controller are developed, evaluated and compared. The results show that the bicycle can ride without human interaction on a bicycle roller in different velocities. Additionally, numerous experiments are conducted in an outdoor environment in several different terrains, where the proposed control structure manages to balance and steer the bicycle.

No description available.

During the last decades, improved sensor and hardware technologies as well as new methods and algorithms have made self-driving vehicles a realistic possibility in the near future. Thanks to this technology enhancement, many leading automotive and technology companies have turned their attention towards developing advanced driver assistance systems (ADAS) and self-driving vehicles. Autonomous vehicles are expected to have their first big impact in closed areas, such as mines, harbors and loading/offloading sites. In such areas, the legal requirements are less restrictive and the surrounding environment is more controlled and predictable compared to urban areas. Expected positive outcomes include increased productivity and safety, reduced emissions and the possibility to relieve the human from performing complex or dangerous tasks. Within these sites, different truck and trailer systems are used to transport materials. These systems are composed of several interconnected modules, and are thus large and highly unstable while reversing. This thesis addresses the problem of designing efficient motion planning and feedback control frameworks for such systems. First, a cascade controller for a reversing truck with a dolly-steered trailer is presented. The unstable modes of the system is stabilized around circular equilibrium configurations using a gain-scheduled linear quadratic (LQ) controller together with a higher-level pure pursuit controller to enable path following of piecewise linear reference paths. The cascade controller is then used within a rapidly-exploring random tree (RRT) framework and the complete motion planning and control framework is demonstrated on a small-scale test vehicle. Second, a path following controller for a reversing truck with a dolly-steered trailer is proposed for the case when the obtained motion plan is kinematically feasible. The control errors of the system are modeled in terms of their deviation from the nominal path and a stabilizing LQ controller with feedforward action is designed based on the linearization of the control error model. Stability of the closed-loop system is proven by combining global optimization, theory from linear differential inclusions and linear matrix inequality techniques. Third, a systematic framework is presented for analyzing stability of the closed-loop system consisting of a controlled vehicle and a feedback controller, executing a motion plan computed by a lattice planner. When this motion planner is considered, it is shown that the closed-loop system can be modeled as a nonlinear hybrid system. Based on this, a novel method is presented for analyzing the behavior of the tracking error, how to design the feedback controller and how to potentially impose constraints on the motion planner in order to guarantee that the tracking error is bounded and decays towards zero. Fourth, a complete motion planning and control solution for a truck with a dolly-steered trailer is presented. A lattice-based motion planner is proposed, where a novel parametrization of the vehicle’s state-space is proposed to improve online planning time. A time-symmetry result is established that enhance the numerical stability of the numerical optimal control solver used for generating the motion primitives. Moreover, a nonlinear observer for state estimation is developed which only utilizes information from sensors that are mounted on the truck, making the system independent of additional trailer sensors. The proposed framework is implemented on a full-scale truck with a dolly-steered trailer and results from a series of field experiments are presented.

With the rapid development of computers and their proven usability in manufacturing environments, simulation-based optimisation has become a recognised tool for proposing near-optimal results related to manufacturing system design and improvement. As a world-leading manufacturer within their field, Volvo GTO in Skövde, Sweden is constantly seeking internal development and has in recent years discovered the possibilities provided by flow simulation. The main aim of this thesis is to provide an optimal buffer size of a new post-assembly and packaging line (Konpack) yet to be constructed. A by-product of the flow simulation optimisation project in form of a flow simulation process evaluation was also requested. The simulation project started with a pre-study including the development of the frame of reference and an analysis of the literature focused on merging Lean philosophy with simulation-based optimisation. The simulation model was built based on both historical and estimated data. The optimisation results showed different buffer size alternatives depending on the throughput to be achieved, these are discussed, and near-optimal solutions presented for decision-making. Additionally, four experiments were carried out, both contributing to the model’s credibility as well as providing new and valuable insight to the stakeholders. The conclusions drawn from the optimisation and experiments indicate that Konpack will be able to meet the established throughput goals, provided that the suggested near-optimal solutions are considered. The experiments also unanimously point to the fact that Konpack has a built-in overcapacity, utilizable by optimising certain suggested input parameters. Additionally, an evaluation of the completeness of the standard simulation process employed by Volvo GTO is provided, concluding that no major changes are needed. Nevertheless, there is always room for improvement. Hence, future work regarding the flow simulation process at Volvo GTO is proposed.

Using robots to navigate through un-mapped environments, specially man-made infrastructures, for the purpose of exploration or inspection is a topic that has gathered a lot of interest in the last years. Micro Aerial Vehicles (MAV's) have the mobility and agility to move quickly and access hard-to-reach areas where ground robots would fail, but using MAV's for that purpose comes with its own set of problems since any collision with the environment results in a crash. The control architecture used in a MAV for such a task needs to perform obstacle avoidance and on-line path-planning in an unknown environment with low computation times as to not lose stability. In this thesis a Non-linear Model Predictive Controller (NMPC) for obstacle avoidance and path-planning on an aerial platform will be established. Included are methods for constraining the available state-space, simulations of various obstacle avoidance scenarios for single and multiple MAVs and experimental validation of the proposed control architecture. The validity of the proposed approach is demonstrated through multiple experimental and simulation results. In these approaches, the positioning information of the obstacles and the MAV are provided by a motion-capture system. The thesis will conclude with the demonstration of an experimental validation of a centralized NMPC for collision avoidance of two MAV's.