Autonomous target search is crucial for deploying Micro Aerial Vehicles (MAVs) in emergency response and rescue missions. Existing approaches either focus on 2D semantic navigation in structured environments – which is less effective in complex 3D settings, or on robotic exploration in cluttered spaces – which often lacks the semantic reasoning needed for efficient target search. This thesis overcomes these limitations by proposing a novel framework that utilizes semantic reasoning to minimize target search and exploration time in unstructured environments using a MAV. Specifically, the open vocabulary inference capabilities of Large Language Models are employed to embed semantic relationships in segmentation images. An active perception pipeline is then developed to guide exploration toward semantically relevant regions of 3D space by biasing frontiers and selecting informative viewpoints. Finally, a combinatorial optimization problem is solved using these viewpoints to create a plan that balances information gain with time costs, facilitating rapid location of the target. Evaluations in complex simulation environments show that the proposed method consistently outperforms baselines by quickly finding the target while maintaining reasonable exploration times. Real-world experiments with a MAV further demonstrate the method's ability to handle practical constraints like limited battery life, small sensor range, and semantic uncertainty.

Learning new concepts is a difficult task for autonomous robots. These robots can adapt to changes in the situations. To adapt to a situation, they should be able to determine the usefulness of objects around them. The usefulness of objects is highly dependent on situational context, making pre-programming of adaptation behaviour to all possible situations difficult. Automatic learning of this information by the robot from its own observations during deployment is a more feasible approach. We encode the usefulness of objects in logical statements that symbolise regularities in observations of objects in a Markov Logic Network. This network forms the basis from which the usefulness of newly observed objects or situations can be inferred by a robot. Each logical statement in the network has an associated weight that indicates how strong the robot believes the statement is true. These weights can be adjusted over time based on new observations made by the robot, strengthening its beliefs or creating a new belief for an unknown object in a process called cumulative learning. We extend the Markov Logic Network framework with a cumulative learning algorithm called MLN-CLA. This algorithm contains several different knowledge updating strategies that handle conflicts. We demonstrate the ability of MLN-CLA to learn the usefulness of unknown objects over time. In addition, we demonstrate the abilities of MLN-CLA to incorporate new logical statements to better capture information about objects and situational contexts. Together, these two abilities enable robots to autonomously adapt to changing situations during deployment.

The offshore industry is adopting Autonomous Underwater Vehicles (AUVs), to decrease the costs associated with surveying underwater sites. Despite requiring less human supervision, AUVs still depend on costly servicing from a vessel, before and after deployment. Combining an AUV with an Unmanned Surface Vessel (USV), a survey vessel that operates without humans on board, could drastically reduce operational costs. This thesis focuses on the crucial requirement of this combination: the autonomous docking of the AUV to the USV.

Current studies focused on the AUV-USV combination lack effective solutions for docking in rough ocean conditions, while this is key for its operational employability. These studies have explored horizontal docking approaches, while an unexplored alternative is a vertical docking approach from below the USV. This approach may experience reduced influence of waves, as the wave disturbance acting on the AUV and the USV are phase synchronized. The objective of this study is to evaluate the viability of vertically docking an AUV to a USV in rough seas, using the Lobster Scout AUV as a case study for an initial performance evaluation.

This study first identified and scoped the key elements for this initial performance evaluation, which included the waves, the USV, the Lobster Scout, the Docking Station (DS), and vertical docking Guidance Navigation and Control (GNC). Furthermore, the scope was limited to the critical docking stage, where the vehicles are within meters of each other. The study used simulation as the primary method to investigate the thesis objective and a 2-Dimensional (2D) model was developed to simulate the system dynamics, which was fitted and verified using a variety of methods, including analytical calculations, mesh convergence studies, experiments, and visual analysis.

To determine the viability of the vertical docking approach of the USV-AUV combination, a probability distribution-based method was developed. Viability was expressed in terms of the maximum operational sea state and the estimated operational up-time. Since docking success under the presence of waves is described probabilistically, the requirement was set that at least 991 out of a 1000 docks should be successful. Furthermore, Monte Carlo simulations were used to obtain the docking performance over a variety of sea states, navigational settings, and different guidance and control methods.

A target state vertical guidance method using way-points was designed with various Proportional-Integral-Derivative (PID) based controllers to guide the Lobster Scout to the DS. It was also determined whether kinematic prediction or polynomial prediction could improve docking performance and the impact of navigation on docking performance was investigated. It was found that the vertical docking approach shows promise for enabling a reliable AUV-USV combination for the Lobster Scout in medium to rough seas, with a significant wave height of 1.65 m and 3.6 m respectively, resulting in an estimated annual operational up-time of 59 % to 94 % in the North Sea. Furthermore, there is potential to achieve even higher sea states with improved GNC methods.

Overall, this study suggests that the vertical docking approach can lead to a viable AUV-USV combination with improved operational employability compared to current docking studies using horizontal approaches, although the author is of the opinion that both approaches require further investigation.

Autonomous robots hold great potential for positive impacts on society by applying them to tasks that are hazardous, repetitive, or complex and difficult for humans to perform. To achieve these tasks, autonomous robots require the ability to perceive environmental changes and create corresponding motion plans, which involve a combination of perception and motion planning techniques. Building a perception pipeline that can detect all relevant details in a dynamic environment is challenging and computationally expensive. To address this issue, the raw data output of a distance-measuring sensor can be used as a perception pipeline directly, transferring most of the computational load to the motion planner. However, most motion planning techniques cannot handle this high computational load. The motion planning technique optimization fabrics offers a promising solution, which utilizes a combination of differential equations to design robot behavior with extremely low computational complexity.

This thesis proposes a method for local motion planning that combines the low computational complexity of optimization fabrics with direct sensor integration. Our goal is to develop a method that enables autonomous robots to perceive and respond to changes in their environment quickly and efficiently. Our method, called direct sensor integrated (DSI) optimization fabrics, utilizes collision-avoidance differential equations generated for each raw data point collected from a distance measuring sensor. We adapted the regular optimization fabrics method to incorporate sensor data directly, and our analytical analysis shows that direct sensor integration does not require scaling adjustments to the regular collision-avoidance differential equations. By combining optimization fabrics with direct sensor integration, DSI optimization fabrics offers a promising solution to local motion planning. This method can enable autonomous robots to handle tasks that are challenging for humans to perform without needing a complex perception pipeline.

We conducted several experiments to assess our method, utilizing a simulated LiDAR-equipped point robot. First, we show empirically that an adjustment is required to properly scale the collision-avoidance differential equations, resulting in similar collision-avoidance behavior across sensor resolution. Second, we demonstrate that DSI optimization fabrics is feasible regarding computational complexity, achieving a frequency of 23 Hz even with the maximum sensor resolution of 2048 LiDAR rays. Third, we demonstrate the effect of varying the sensor resolution on performance in multiple goal-reaching scenarios. We measure performance by monitoring the time-to-goal, total path length, minimum clearance from obstacles, and success rates. Fourth, we compare our method to regular optimization fabrics with a simulated perception pipeline in a scenario with static obstacles and a scenario with dynamic obstacles. In both scenarios, we show comparable performance regarding time-to-goal, total path length, minimum clearance from obstacles, and success rates. Finally, we showcase a real-world application of our method.

The simultaneous rapidly increasing demand for home delivery of goods and on-demand expectancy of customers over the past years leaves a tough challenge for the logistical branch. They have to keep up with this increasing demand and simultaneously they are obliged to satisfy consumer service level demands to preserve their customers. On the other hand, as economic goals drive these businesses, they are prompted to operate cost-effectively. As a result, the fleet, deployed to execute the last-mile delivery, should meet both the requirement of cost-efficiency as well as the requirement for meeting consumer service level demands. This raises the question of how to efficiently design a fleet for last-mile on-demand logistics. For a fleet to be able to operate cost-efficiently, the fleet design decisions are required to take both fixed and variable costs into account. As such, the fleet design decisions need to include the consideration of the size of the fleet as well as the distance the vehicles travel on daily basis. Therefore, the goal of this thesis is to develop a novel method for fleet design for last-mile on-demand logistics. This work contributes by being the first to investigate methods for doing fleet design specifically for last-mile on-demand logistics considering multiple depots and variable pick-up locations. The purpose of the method is to determine the operational plans of the individual vehicles, the number of vehicles needed throughout a certain time period, the pick-up locations for all orders and the total distance travelled by the full fleet of vehicles. The proposed method builds upon established fleet design methods for ride-sharing taxi problems. The optimization method is adapted for last-mile on-demand logistics, yielding the required number of vehicles and their individual operational plans. The input of the system is a set of trips, which represent a path of a single vehicle to deliver one or multiple orders from a depot. Connecting two trips, which is called chaining, has the benefit of reducing the number of vehicles used, as chained trips are served by a single vehicle. Additionally, from multiple available depots where orders can be picked up, the method determines the best depot per order. This part of the method is called depot re-assignment. Furthermore, the fleet design problem is modelled as a multi-objective optimisation problem to find the trade-off between fleet size and the total distance the vehicles travel. Three different modelled datasets, each containing 10.000 order requests in the city centre of Amsterdam, are used to prove the value of the given method. A comparison between the method with and without depot re-assignment is made, to prove the value of the given addition of depot re-assignment. It is proven that depot re-assignment is valuable as it decreases or retains the fleet size for all test cases. The experiments conducted show that a significant decrease of the required fleet size can be established by a minor increase in total travelled distance. Furthermore, the optimal trade-off between the fleet size and the total distance travelled can be determined for a specific operation with the knowledge of operational costs for that operation.

Quay walls are important structures that keep the water in harbours and canals within their bounds, and accommodate large infrastructure like roads or ship-handling structures on top of them. Due to their importance it is critical that they do not fail or collapse. Inspections are done to prevent this, as a better knowledge of the state of quay walls can help in predicting future behaviour of the quay walls. However, current inspection methods are not satisfactory and can be improved upon. This thesis proposes the Raybot. This is an Autonomous Underwater Vehicle that can move along the quay walls like rays swim along the walls of their aquariums. The design of the Raybot is presented, which keeps in mind the requirements that follow logically from the quay wall inspection mission. This resulted in a rectangular robot with an outward modular frame. The thrusters, on-board computers, and sensors can be mounted on the inside. Next, a Model Predictive Controller is proposed for the motion control of the Raybot. For this, the extensive kinematics and kinetics of the Fossen model are explained. The parameters of this Fossen model are estimated for the Raybot using various methods. A Model Predictive Controller is formulated for Autonomous Underwater Vehicles and implemented in the Robot Operating System 2. A physical and visual representative simulation environment is set up, which can simulate the motion of the Raybot. This is used to assess the path tracking behaviour of the Raybot. The Model Predictive Controller is compared to a cascaded PID controller by path tracking of a zig-zag path along the quay wall. An analysis of the tuning parameters of the Model Predictive Controller is presented. The Model Predictive Controller outperforms the cascaded PID controller on both the Mean Squared Error of the path tracking error and the completion time of the inspection mission, whilst adhering to constraints set by the minimum and maximum velocity of the Raybot and the minimum and maximum thruster inputs. This improved performance leads to a higher quality of the inspections, as they can be done more up-close, as well as a higher quantity of the inspections, as more inspections can be done in the same time. For future work it is recommended to estimate some parameters of the model on a real robot, as well as testing the Model Predictive Controller on a real robot in a tank. This will showcase the performance of the motion controller in an even more realistic scenario.

In this work, we address the challenges of employing robots in the Search-and-Rescue (SAR) domain, where they can benefit rescue workers to quickly obtain Situational Awareness (SA). Missions with autonomous mobile robots are heavily dependent on environmental representations. Representations have been steadily increasing in the richness that they can capture, in addition to geometry we can now represent objects and their interrelations. These richer environmental representations, such as the 3D scene graph, have provided an opportunity to integrate representations with prior knowledge to make them more actionable and improve the SA they provide. The use of such representations for planning is limited. In previous work, the main approach was to augment and extend the scene graph to enable traditional symbolic planning. The limitations of these methods are that they are offline, scale poorly, and require full observability, making them unsuitable for the SAR domain. The main contributions of this work are as follows. First, we propose the behavior-oriented situational graph, a data structure that integrates data-driven perception with prior knowledge following a novel situational affordance schema. This schema connects situations with robot behaviors and mission objectives, allowing for autonomous mission planning. Second, we propose an efficient method to obtain task utilities from the proposed behavior-oriented situational graph through planning. Finally, we propose an exploration component to discover and select tasks online in dynamic environments that are potentially partially observable. This work is evaluated in several simulation scenarios, showing improved efficiency in mission completion compared to offline methods for the specific SAR domain. Finally, the methods are implemented and tested in a real-world scenario using a mobile Spot robot, showing its effectiveness in practice.

Grasping objects in a smooth humanlike motion, instead of the more typical pick-and-place approach, includes multiple aspects that need to be performed correctly for a successful grasp. These aspects involve moving the end-effector such that its surface makes and retains contact with the object while also coordinating the movement of the gripper to securely grasp the object. This work investigates how the intricate task of grasping may be learned from humans based on kinesthetic demonstrations. Due to the complexity of the task, these demonstrations are often slow and even slightly flawed, particularly at moments when multiple aspects (i.e. end-effector movement, orientation and gripper width) have to be demonstrated at once. Rather than training a person to provide faster demonstrations, non-expert users are provided with the ability to interactively modify the dynamics of their initial demonstration through teleoperated corrective feedback. This in turn allows them to teach motions outside of their own physical capabilities. In the end, the goal is to obtain a faster but reliable execution of the task.
The presented framework learns the desired movement dynamics based on the current Cartesian position with Gaussian Processes, resulting in a reactive, time-invariant policy. Using Gaussian Processes also allows online interactive corrections and active disturbance rejection through epistemic uncertainty minimization.
The experimental evaluation of the framework is carried out on a Franka-Emika Panda. Tests were performed to determine i) the framework's effectiveness in successfully learning how to grasp an object quickly, ii) ease of policy correction to environmental changes (i.e. different object shapes and mass), and iii) the framework’s usability for non-expert users.

As e-commerce continues to grow, warehouses face the challenge of keeping up with product flows. Vanderlande provides a pick-and-place robot (SIR) that helps in improving capacity. However, SIR's capacity is limited by its vacuum gripper and its inability to solve failures. The aim of this thesis is to improve SIR by adding a second gripper and reasoning capabilities. To this end, a grasp synthesis approach is developed that complements the vacuum gripper, based on existing approaches in literature and the devices available to SIR. Furthermore, an ontology is developed according to Methontology that allows the robot to reason about item properties and failures, while providing operators with insight into its thought process. The improved system is referred to as RSIR.
The grasp synthesis approach, applicable to parallel-jaw grippers, synthesizes grasps by detecting parallel edges in depth data. Tests on real depth data showed that the approach synthesized reasonable grasps in reasonable time for a variety of items, some of which are not graspable by SIR's vacuum gripper. It did not need a segmentation of the items to do so. Furthermore, RSIR was tested on a virtual robotic set-up. The results showed that RSIR reasoned how best to complete a task when provided with information about a robot's components and possible item properties. RSIR also drew conclusions based on the execution of the task, and reasoned what actions should be performed to prevent failures from re-occurring. The above results show RSIR's potential for autonomously estimating item properties and selecting the best approach for grasping an item. In practice, this decreases downtime and thus improves capacity. Because RSIR is designed to be expandable to related tasks, its benefits could be useful in more of Vanderlande's applications than just the SIR system.

The advancements in driver assistance systems in cars and the development of autonomous vehicles require more states of the car to be known than the conventional sensors can accurately determine. Load sensing bearings (LSBs) are under development that measure the wheel forces under dynamic load. Through sensor fusion, the addition of this sensor can increase the robustness and accuracy of the state estimators used in cars. An algorithm uses the strain measured on the surface of an instrumented LSB to calculate the forces on the bearing. The development of LSBs requires a tool that simulates the strain signals in a fast and accurate manner to gain insight into the bearing behavior.
Due to the complex nonlinear behavior of the bearing, there is not a fast, straightforward tool available that estimates its strains. This thesis presents a methodology that uses finite element analyses (FEA) to construct a model that calculates the strain in the outer ring of a wheel bearing for any given load. The FEA consists of a linear elastic model of the outer ring and a single loaded ball that is modeled by a Hertzian contact. Multiple simulations are done for different positions of the ball, such that a rotating bearing can be approximated. A nonlinear analytical bearing model in conjunction with a, from FEA constructed, outer ring flexibility model calculates the load on each bearing ball. A strain model, also built from the FEA, uses these loads to calculate the strain on the outer ring.
The simulated strains are validated with experiments performed on a bearing test rig of SKF. Measurements from a with strain gauges instrumented bearing show that the model predicts the observed behavior in the signals. Analyses of the simulated and measured signals in the frequency domain show a difference in gain and offset, which can be calibrated. Unexpected discrepancies are observed within the measured signals of symmetrically placed sensors on the LSB, which should give identical signals, that are likely caused by a distortion of the shape of the outer ring introduced by the manufacturing process or installation of the bearing.
The scope for future work should focus on further validation of the model and developing a calibration method to increase the accuracy. It is believed that the uncertainties in the model can be summarized into a small set of parameters that can be calibrated for a specific instrumented LSB with only a few measurements. Once accurate strains can be simulated, the model could be utilized in a state observer to convert actual strain measurements into loads, and it is possible to use it to optimize the design of the LSB and the location of the strain gauges.

Simultaneous Localization And Mapping (SLAM) algorithms provide accurate localization for autonomous vehicles and provide essential information for the path planning module. However, SLAM algorithms as- sume a static environment in order to estimate a location. This assumption influences the pose estimation in dynamic urban environments. The impact of this assumption on day-to-day scenarios of an intelligent vehicle is unknown. A deeper understanding on the effect of dynamic scenarios in an urban environment could lead to simple and robust solutions for SLAM algorithms in intelligent vehicles. The objective of this research is to develop a methodology that isolates the effect of an urban dynamic environment on the per- formance of a SLAM algorithm. This requires constant environment conditions including constant weather conditions, lighting conditions and identical trajectories over time. The methodology is tested with a stereo feature based V-SLAM algorithm called ORB SLAM [19], which illustrates the in-depth analysis that is possi- ble with this experiment. The main research question is: How does a dynamic urban environment influence the pose estimation accuracy of stereo ORB SLAM? Two specific dynamic scenarios are designed to represent a dynamic urban environment: driving behind another vehicle and vehicles approaching on the other side of the road. On these scenarios, an in-depth anal- ysis of ORB SLAM is performed to observe how the algorithm’s design influences the robustness to a dynamic environment. Functions within the algorithm are bypassed to analyze the effect on the performance. Specifi- cally, the place recognition function and map point filtering function are bypassed. The analysis proofs which functions assist in the overall robustness to a dynamic environment. Moreover, an analysis is performed of the algorithm in localization mode to research the effect of utilizing maps that were created under different conditions. The knowledge gained from the full analysis can be utilized to improve other V-SLAM algorithms. The experiment is performed in CARLA [6], an open source simulator. CARLA provides an elaborate sen- sor suite which support multiple camera setups and LIDAR sensors. Furthermore, the simulator provides free maps which represent realistic urban environments and allows for easy and accurate access to the ground truth position. A setup is designed with the simulator that allows complete isolation of the effect of a dy- namic environment. The setup allows full control of lighting conditions, weather conditions and allows iden- tical trajectories over time in different dynamic scenarios. Each scenario is simulated over several different trajectories in which the camera images are converted to rosbags. Each variation of the ORB SLAM algorithm is tested on the produced rosbags. The resulting pose estimations in dynamic conditions are compared to the pose estimations made during static conditions to analyze the effect of dynamic scenarios on the perfor- mance of the algorithm. The method successfully isolated the effect of a dynamic environment on the performance of stereo ORB SLAM. It allows for a detailed analysis which aids in finding the source of performance differences. In general, stereo ORB SLAM displays robust behavior to a dynamic environment. The experiment shows that the algo- rithm is sensitive to false relocalization when the stereo camera setup is driving 10 meters behind another vehicle for a long period of time. During these conditions, ORB SLAM cannot provide accurate pose esti- mations even when the place recognition module is deactivated. Furthermore, the map point filtering does increase the robustness in certain dynamic scenarios. Finally, the data suggests that utilizing maps created in different conditions does influence the pose estimation in localization mode. However, more data is needed to confirm these results. The methodology has proven its value for in depth analysis of robustness to an urban dynamic environ- ment for a SLAM algorithm. This experiment is not limited to ORB SLAM but could be utilized for other monocular and stereo V-SLAM methods, as well as LIDAR based methods. New solutions can be developed to increase robustness to a dynamic environment and tested on the same rosbags. This methodology could be an important tool for the development SLAM algorithms for intelligent vehicles.

This research considers using center of mass (CoM) height variation as an input for balance control on a humanoid robot. Traditional balance strategies for humanoid robots are taking a step, control of the center of pressure (CoP) location, a result of the ‘ankle strategy’, and changing the angular momentum about the CoM, for example by a ‘hip strategy’. For humanoid robots, a common assumption behind these strategies is that the CoM height is predefined. However, CoM height changes can be used as an input for balance control, as for example can be observed during the landing of an athlete after a long jump. The first contributions of this work are bounds on the initial states for the variable height inverted pendulum (VHIP) from which convergence is possible to a stopped state, also known as capture regions. First, only a unilateral contact constraint is considered; negative CoM acceleration cannot be smaller than gravitational acceleration. Second, CoM height constraints are added to the model, after which a capture region can still be computed in closed-form. Third, vertical force constraints are added, after which capture regions are computed numerically using a bang-bang control law. The last capture region bridges the transition to the applied part of this work. The second contribution is a control law on vertical acceleration, suitable for application on a humanoid robot using a momentum-based control framework. Push recovery is tested on NASA’s Valkyrie humanoid robot while the robot is standing. In simulation, an increase in recoverable push of 9% can be observed when comparing to a controller that only uses CoP, when pushing the back of the robot. On hardware, an average increase of 7% can be observed for this push direction using a load sensor. Additionally, tests are conducted on hardware on Boston Dynamics’ Atlas using a medicine ball on a rope, but no improvement in recovery is observed. The control method for standing push recovery is also extended for use while the robot is walking. For Valkyrie in simulation, recovery improved the most compared with a predefined height approach for a push applied in the first part of the single support state for rear and frontal push directions. Additionally, a hardware result on Atlas while walking is briefly presented.