As self-driving vehicles progress toward real-world deployment, efficient and reliable motion planning in dynamic multi-agent environments becomes increasingly essential. This work addresses this challenge by advancing the field of nonlinear distributed model predictive control (NMPC) for autonomous multi-vehicle coordination. The approach focuses on complex, interaction-dense scenarios where space is limited. To model vehicle shapes and collision boundaries efficiently it uses polytopic set representations to approximate vehicle geometries and collision boundaries instead of conventional shapes such as ellipsoids.

We propose a novel distributed NMPC approach for navigation in tight environments. The goal is to enable coordinated motion planning for multiple autonomous vehicles in dense traffic scenarios. This can be easily modelled with centralised formulations, however, considering the scale of the road network they become computationally intractable as the number of agents grows. In distributed approaches instead each vehicle solves its local part of the problem which scales linearly and is therefore better suited for these kind of environments. So building on an existing distributed method we enhance its structure to improve scalability, safety, and performance in cooperative autonomous driving tasks. The research is guided by three objectives: (RO1) to reproduce a known distributed baseline algorithm using the DART (Delft’s Autonomous-driving Robotic Testbed) vehicle model \cite{lyons_dart_2024}, the Model Predictive Contouring Control (MPCC) \cite{lam_model_2010} tracking objective, and the ACADOS solver framework \cite{verschueren_acados_2020}; (RO2) to develop a novel distributed MPC-based algorithm that improves inter-vehicle spacing and tracking performance while generalizing beyond predefined reference trajectories; and (RO3) to prepare the algorithm for validation on a real-world robotic testbed.

The primary contributions of this work include: (C1) successful reproduction of the distributed baseline using open-source tools and realistic vehicle modelling; (C2) development of two enhanced distributed algorithms, Distributed Model Predictive Contouring Control with Relaxed Collision Avoidance (DMPC-RCA) and DMPC-RCA with consensus (DMPC-RCA-C), that demonstrate superior performance in tracking accuracy and safety margins; and (C3) integration of these algorithms with the DART hardware platform for future experimental validation. The repository of these approaches can be found in https://github.com/HubertVisser/multi-vehicle-coordination-algorithm.git

Performance is evaluated in two representative scenarios, a merging situation and a T-junction, with the centralised NMPC approach serving as the performance benchmark. To evaluate the performance of the proposed designs, three key metrics are used: accumulated tracking cost, computation time, and minimum inter-agent distance. The results show that both proposed methods achieve tracking costs comparable to the centralised controller, while significantly outperforming the distributed baseline method. Notably, the inclusion of a consensus term yields no substantial improvement in performance over the non-consensus version.

To conclude, the proposed approaches offer strong potential for scalable, safe, and efficient multi-agent motion planning, moving one step closer to the deployment of fully cooperative autonomous driving on public roads.

Github repository for the Master Thesis - https://github.com/HubertVisser/multi-vehicle-coordination-algorithm.git

This thesis proposes a study towards the application of imitation learning (IL) algorithms Action Chunking Transformer (ACT) and diffusion policy as an autonomous surface vessel (ASV) path planner in complex marine environments. Rationale for conduct- ing this research are the ubiquitous limitations regarding fine tuning cost- and or reward functions of conventional state of the art algorithms for ASV navigation in complex environments. In this study we trained both algorithm types on data sources collected from a basic 2D-grid sim- ulator and a more realistic GAZEBO simulator. Subse- quently we evaluated both algorithm’s performances in each respective simulator in terms of the success rate for a standard navigation task. We found relatively high suc- cess rates for the 2D-grid simulator (0.98 for ACT and 0.53 for diffusion policy). For the GAZEBO simulator, we found poor performance for ACT (0.0 success rate) and for diffusion policy we could not establish the perfor- mance due to hardware limitations. For future work, the capabilities of both models could be investigated further by trying to bridge the gap between the simple 2D simula- tor and the GAZEBO simulator. Mainly the effect of task complexity and the quality and quantity of data used for training the models on the performances of the models can be investigated in these future work studies.

The rapid increase in drone threats has created an urgent need for practical counter-drone systems for safety and defence purposes. Interceptor drones represent a promising and cost-effective solution for removing unwanted aerial vehicles. However, classical guidance laws such as Proportional Navigation (PN) are not well adapted to highly agile targets under realistic perception constraints. In parallel, deep reinforcement learning (DRL) has demonstrated exceptional performance in high-speed quadcopter control tasks, motivating its consideration for interception applications. A major challenge for such learning-based controllers lies in sim-to-real transfer. This work addresses this challenge through extensive domain randomization and explicit modelling of realistic perception constraints, including sensor noise, limited field of view, sensing ranges, and sensor update rates. A framework is proposed that trains Single Rotor Thrust (SRT) DRL policies for quadcopters and evaluates them against a classical PN implementation within a physics-based simulator. The resulting controllers generalise across platform variations and outperform classical guidance under modelled perception limitations, demonstrating improved robustness and transferability.

Github Repository of the code used for the thesis: https://github.com/maximecapelle/quadintercept_drl

Mechanical tracking systems offer potential advantages over optical and electromagnetic alternatives for surgical navigation, including freedom from line-of-sight constraints, immunity to metallic interference, and high-frequency position feedback. However, passive mechanical tracker arms lack the joint locking capability required for traditional laser tracker calibration, and existing artefact-based methods fail to provide the complete six-degree-of-freedom constraints necessary for accurate kinematic parameter identification. This thesis presents a novel artefact-based calibration methodology specifically designed for passive mechanical tracking arms used in surgical navigation applications. The methodology employs a laser-certified steel calibration artefact featuring 15 asymmetric pillars that provide complete 6-DoF constraint through single contact points. Unlike traditional sphere-based methods that only constrain position, the asymmetric plug design mechanically prevents rotation, enabling simultaneous identification of all kinematic parameters without sequential joint locking. The artefact was precision-machined and certified using a Leica AT960 laser tracker to ±0.001 mm uncertainty, providing metrological traceability one order of magnitude better than the target accuracy. Applied to a custombuilt six-degree-of-freedom Surgical Tracker Arm (SUTA), the calibration methodology achieved 1.01 mm RMS accuracy across 525 independent validation measurements, representing an 81.5% improvement from the 5.45 mm RMS baseline configuration. Dataset optimization analysis revealed that six measurement positions achieve near-optimal accuracy (1.09 mm RMS) while maintaining calibration efficiency, with marginal improvements beyond this point. The research identified mechanical interface compliance as the dominant error source, with clearance in the end-effector-pillar interface contributing approximately 0.5 mm to the total error. While the achieved accuracy does not yet match the 0.3-0.5 mm performance of state-of-the-art optical systems, the methodology successfully demonstrates that passive mechanical trackers can be calibrated to millimeter-level precision with a calibration artefact. The approach establishes a practical framework for calibrating passive tracking systems and identifies clear paths toward sub-millimeter accuracy through targeted hardware improvements, particularly in interface design and consistent seating mechanisms.

Reinforcement learning has emerged as a promising approach for enabling robots to learn from interactions with their environments, without relying on predefined behaviors. However, robots face significant challenges when learning directly from real-world interactions. Real-world learning is time-consuming and resource-intensive, often requiring extensive data collection over long periods. Additionally, the risks involved in trial-and-error learning in physical settings are high, as faulty policies can lead to safety issues or system damage. Simulations offer a safer and more efficient alternative, allowing robots to learn in simulated environments at faster-than-real-time speeds. Despite these benefits, simulations often serve as imperfect approximations of reality. As a result, robots may learn behaviors that exploit simulation-specific quirks, which may not perform well in real-world settings, creating difficulties in transferring learned behaviors from simulation to real environments—a challenge known as the sim-to-real gap. Several factors contribute to the sim-to-real gap, such as unmodeled physical phenomena like friction and deformation, and the asynchronous nature of real-world systems that simulations often fail to capture accurately. Additionally, using separate software stacks for simulation and deployment can unintentionally lead to discrepancies. Finally, simulating at faster-than-real-time speeds with asynchronous frameworks that distribute computation across multiple cores may also introduce inaccuracies without proper synchronization.

This thesis focuses on improving simulation tools and methodologies to enhance the efficiency and effectiveness of learning-based approaches in robotics. The work addresses key trade-offs between flexibility, speed, and accuracy in robotic simulations, which are critical for successfully transferring learned policies from simulation to real-world environments. Additionally, it introduces a strategy to improve resilience, ensuring that learned behaviors are robust to irrelevant and unknown dynamics.
By tackling these challenges, this thesis provides insights into the design of effective robotic simulators and presents contributions that help bridge the gap between simulated and real-world robotic learning.

This thesis focuses on enabling safe and reliable navigation for Autonomous Surface Vessels (ASVs) operating in complex and mixed-traffic maritime environments. It presents a suite of motion planning and control algorithms that ensure fault tolerance and compliance with maritime traffic rules, even under uncertainty and component failures. Key contributions include a Model Predictive Contouring Control (MPCC) method that formalizes COLREGs compliance, a model-based fault diagnosis framework using residual analysis, a robust Set-Membership Estimation (SME) approach for fault parameter identification, and a Robust Adaptive Model Predictive Control (RAMPC) scheme that integrates fault information into trajectory optimization. Validated through extensive simulations and implemented in ROS, the proposed framework demonstrates robust performance in dynamic and uncertain conditions, laying the groundwork for real-world deployment of autonomous maritime systems.

Public safety and emergency response agencies increasingly consider the deployment of mobile robots as mounting climate-related disasters and security challenges place human personnel at higher risk and stress. Mobile robots, such as drones, are a promising strategy to respond to these challenges: They can navigate difficult, hazardous terrain, gather real-time situational data, and conduct search or reconnaissance tasks without putting humans at direct risk. However, the currently practiced teleoperation of robots is challenging for such complex missions since the simultaneous navigation, situation assessment, and search tasks can overload human cognitive abilities. Therefore, autonomous planning and decision-making algorithms are required to enable robots to explore and search unknown environments for targets such as missing persons or hazardous materials.

Moving towards this goal, this thesis addresses two core problems. First, local motion planning must carefully account for information gained from sensor observations as well as collision avoidance and the robot’s dynamics while moving through cluttered, unknown areas. Second, global exploration planning must strategically select where in the environment to explore to find the target quickly - especially when the environment is large or complex. Given that human operators often possess semantic knowledge about likely target locations, we hypothesize that incorporating such guidance by observed semantic features (e.g., object or room types) into the exploration planning is crucial for time-efficient autonomous search. We address these two core problems by making the following contributions.

The first contribution of the thesis is an informative local motion planning approach
that generates safe, collision-free trajectories around obstacles while minimizing uncertainty about the target locations. The critical challenge is to achieve computationally efficient planning of trajectories that maximize information gain under the robot’s kinodynamic constraints. In the proposed approach, a model predictive control (MPC) motion planner is guided by a learned viewpoint policy. The policy is trained via deep reinforcement learning (DRL) to maximize long-term information gain by providing a local subgoal to the MPC. The MPC follows the subgoal and ensures that the motion plan remains feasible and collision-free. Therefore, the robot can rapidly replan safe and informative local trajectories online. Simulation experiments demonstrate that the method achieves competitive performance in locating targets compared to a computationally expensive state-of-the-art planner using Monte Carlo Tree Search (MCTS), while allowing for significantly faster execution and replanning.

While local informative planning is crucial for exploring cluttered spaces, it often be-
haves myopically and inefficiently with respect to large and complex environments. Therefore, the second contribution introduces a global target search planner that balances directed search towards semantically promising areas with complete search space coverage. This planner extends the idea of frontier exploration - focusing observations on the boundaries between explored and unexplored regions - to target search, where different frontiers are assigned a semantic priority. This priority represents the semantic relationships between the target and nearby objects. To minimize target search time, the target search planner schedules high-priority frontiers earlier by solving a custom combinatorial optimization problem to determine the visitation order. By integrating coverage gains into the frontier priorities, the planner ensures that the robot explores the environment efficiently while focusing on semantically relevant areas. We demonstrate this approach in two studies outlined below.

Large, high-quality datasets for learning target-specific semantic relationships are
scarce in many real-world scenarios, especially in search and rescue. The third contribution addresses this limitation by proposing a method to learn semantic priority models from expert feedback. Rather than collecting massive amounts of labeled data, the approach exploits an expert operator’s sparse guidance inputs in a few target search scenarios. This expert guidance selects a frontier to explore next, which is stored in a training dataset together with the frontier’s semantic features. An expert model is then trained to approximate a priority function that predicts how relevant each frontier is for the expert. By incorporating this learned priority function into the global target search planner, the robot can autonomously prioritize semantically relevant areas according to the expert’s semantic knowledge. Experiments show that using a small number of expert demonstrations is sufficient for the robot to significantly improve its search efficiency and reduce travel distance until the target is found.

Lastly, the thesis extends semantic target search to three-dimensional environments
by integrating it into a 3D planning pipeline for micro aerial vehicles (MAV). The pipeline first detects objects in the environment using onboard vision and associates them with priority values computed from pre-trained large language model (LLM) embeddings. These priorities are then propagated into frontiers in a 3D voxel map, indicating frontier regions that are most likely to contain the target. This enables the evaluation of frontier viewpoints for their information gain that accounts for both semantic priority and volumetric coverage. The viewpoint gains are then used in the combinatorial target search planner to prioritize the viewpoints that most likely lead to the target while ensuring efficient coverage of the environment. By integrating the MAV’s kinodynamic constraints into the planning costs, the system ensures smooth, feasible trajectories in real-time. Simulation studies reveal that semantically guided exploration leads to faster and more reliable target discovery than different purely coverage-based exploration baselines. Experiments with a real MAV in the lab confirm the approach’s ability to autonomously navigate an MAV through complex 3D environments to a target, exploiting semantic cues, maximizing
coverage, and avoiding collisions.

In summary, this thesis demonstrates how planning and learning techniques can be
combined for autonomous target search and exploration. These techniques enable mobile robots to navigate unknown environments efficiently and safely while searching for targets and collecting required information. Crucially, our proposed method for semantically guided frontier planning bridges the gap between recent learning-based navigation approaches and established planning-based approaches suitable for real-world robotic systems. By integrating semantic knowledge into robotic exploration, the proposed methods can reduce human operator cognitive load and, therefore, facilitate robot deployment in scenarios such as search and rescue or reconnaissance missions.

This thesis presents a COLREG-compliant collision avoidance and detection algorithm for the safe navigation of autonomous surface vessels. The method builds upon the Velocity Obstacle algorithm and implements the concept of the ship domain and the ship arena into the framework. The International Regulations for Preventing Collision at Sea – the COLREGs – are analyzed individually and incorporated in the designed collision avoidance algorithm. The COLREGs were used to determine the preferred passing side and optimized motion for every possible encounter situation with an obstacle vessel. The ship domain and arena help to perform proactive alterations when required, improving on the apparent shortcomings of the VO algorithm. This framework leads to a more proactive collision avoidance algorithm that guarantees COLREG compliance and safe navigation in mixed-traffic environments. The robustness and wide adaptability of the developed algorithm are validated in different encounter situations, ranging from vessel-to-vessel batch simulations to complex, multi-vessel scenarios.

This work introduces a novel training strategy for Gaussian Process (GP) models aimed at improving their predictive accuracy and uncertainty quantification capabilities over extended prediction horizons. This improvement is highly relevant for applications in model predictive control (MPC) in the autonomous driving domain. Learning-based MPC strategies typically rely on standard physics-based models augmented with GP models to account for residual nonlinearities and uncertainties not captured by the former. Nonetheless, these conventional approaches often struggle with long-term prediction accuracy, especially when faced with out-of-distribution scenarios, a phenomenon where the model encounters data points that are significantly divergent from the training set. To address these challenges, a multi-step Gaussian process training framework is proposed. This framework yields a GP model capable of making accurate long-term predictions, i.e. a multi-step Gaussian Process (MSGP) model. It achieves this by integrating the simulation of future dynamics into the training process, allowing for the model’s kernel parameters to be tuned toward long-term dynamics. As a result, the MSGP model not only demonstrated the ability to make more stable and accurate long-term dynamic predictions but also with greater confidence. The efficacy of the multistep training framework is shown by the significant improvements in long-horizon dynamics predictions by the MSGP model, achieving an average 19% reduction in mean error and a 90% reduction in variance compared to the standard GP model. Moreover, the efficacy of the MSGP model is further confirmed through its application in a Multi-Step Gaussian Process-based Model Predictive Contouring Controller (MSGP-MPCC), which outperforms a traditional GP-based MPCC (GP-MPCC) baseline controller in lap time and reliability, achieving a 100% success rate in completing laps across ten consecutive simulations without crashing.

Monitoring wildfires using multiple unmanned aerial vehicles (UAVs) is essential for timely intervention and management of the fire while minimising risks to human lives. A research gap was identified for practical UAV solutions integrating critical features, such as localisation, fire safety and battery management. This thesis presents a novel rule-based algorithm designed for effective wildfire monitoring using cooperative UAVs. A simulation environment with a wildfire spread model was developed to test and evaluate various strategies under dynamic conditions. The proposed algorithm operates in two phases: first, locating the fire and then monitoring its progression. Key features include a heat avoidance mechanism to maintain UAV safety and recharging schedules to maximise operational uptime. A fire detection map is centrally created using the infrared data from the UAVs. The convex hull around the fire serves as the basis for path planning to track the fire perimeter. The best strategy is determined and optimised through a series of experiments in the simulation environment. The overall best-performing algorithm makes navigational decisions based on the importance of different sections of the fire perimeter. The importance of the perimeter is constantly updated based on the fire spread, visit times, and distances of the UAVs. The assignment of various UAV tasks was reviewed to enhance UAV cooperation, which showed performance benefits across different scenarios. Other key findings
include the recommendation of a small fleet size of up to six UAVs, with a flight speed of 10 m/s at an altitude of 100 meters, balancing costs, performance and safety. The algorithm demonstrated performance equal to state-of-the-art reinforcement learning techniques
while offering advantages in explainability. Additionally, the algorithm has been successfully validated in a lab environment, demonstrating its potential as a practical and cost-effective solution for wildfire monitoring. This work brings a fire monitoring system closer to real-world implementation and will possibly help fight wildfires effectively.

Logistics and transportation can greatly benefit from the use of autonomous robots, such as self-driving vehicles. Robots can help to move goods or people without human supervision. One of the main components that enable autonomous navigation among humans is motion planning. Motion planning is responsible for computing a collision-free trajectory that moves the robot to its destination, based on the perceived information. The motion planner should be efficient, robust, and safe. This thesis contributes towards this goal by investigating the design of motion planning algorithms for autonomous navigation of mobile robots near humans.

Traditional motion planners for dynamic environments have two key limitations that this thesis aims to address. First, they assume that their model of dynamic obstacles (e.g., humans) is exactly correct, capturing it with a single deterministic prediction. In practice, the robot cannot observe human intentions and must account for its uncertainty about the human's future behavior. Second, motion planners usually compute a single trajectory around an obstacle as a result of previously taken decisions without exploring alternative options. They react slowly or even fail to find a solution when unpredicted changes make this path undesirable. This results in poor planning performance in dynamic environments.

The goal of this thesis is to develop motion planners that account for the uncertainty of human motion predictions and that are consistent and robust in their decision-making in order to deal with unpredicted changes in dynamic environments. To accomplish this goal, this thesis proposes two motion planning frameworks: scenario-based and topology-driven trajectory optimization.

The first contribution of this thesis is Scenario-based Model Predictive Contouring Control (S-MPCC), a real-time capable probabilistic planning framework that incorporates any uncertainty associated with the motion predictions of dynamic obstacles. Contrary to existing methods that only account for small variations around a single predicted trajectory (unimodal uncertainty), the proposed planner accounts for multiple possible trajectories (multi-modal uncertainty). The planner therefore safely accounts for several outcomes, for instance, to express that a pedestrian may or may not cross in front of the robot.

S-MPCC bounds the probability of collision in each time step with all obstacles through Chance-Constrained Optimization (CCO). The CCO is reformulated as an optimization without uncertainty by sampling trajectories from the predicted distribution, known as scenarios. Each scenario represents a possible position of all obstacles in one time step, and the planner avoids collisions with all scenarios. This Scenario Program (SP), through a tailored linearization, can be solved efficiently online. S-MPCC therefore plans probabilistic safe trajectories independent of the underlying distribution of the uncertainty.

S-MPCC considers the probability of collision separately for each time instance in the planned trajectory. The second contribution of this thesis, Safe Horizon Model Predictive Control (SH-MPC), builds on S-MPCC to constrain the joint probability of collision with all obstacles over the duration of the planned trajectory. Existing methods that separately constrain the probability of collision in each time step (temporal marginal) and with each obstacle (obstacle marginal) lead to overly cautious motion planning when safety constraints are enforced. SH-MPC formulates a single chance constraint to bound the overall probability of collision. This CCO is reformulated as an SP where each scenario represents a possible trajectory for all obstacles. To certify the joint probability of collision with the SP, the number of scenarios that affect the motion plan needs to be identified. SH-MPC estimates this quantity at a negligible computational cost during optimization. Consequently, SH-MPC plans trajectories in real-time under generic uncertainties that are less cautious than existing methods without compromising on safety.

The probabilistic safety of S-MPCC and SH-MPC is linked to the underlying accuracy of the prediction model of the obstacles that provide the scenarios. As a third contribution, a joint prediction and planning framework, Partitioned Scenario Replay (PSR), is proposed that replays past observations of human motion as scenarios for scenario-based planning. PSR does not fit a distribution on observed data but directly uses the data as empirical evidence of the underlying uncertainty and thereby provides a real-world safety guarantee.

A key limitation of the developed scenario-based planners and other optimization-based planners is that they locally refine an initial trajectory. This initial trajectory largely determines the quality of the final trajectory, while it does not consider other options. The fourth contribution of this thesis is Topology-driven Model Predictive Control (T-MPC) that concurrently optimizes trajectories, each attempting a different way to pass the obstacles. T-MPC is composed of a guidance planner and several parallel local planners. The guidance planner identifies guidance trajectories for several distinct maneuvers, relying on results from topology to distinguish trajectories. Each local planner is composed of an existing optimization-based planner (e.g., a scenario-based planner) and an additional set of constraints that are derived from one of the guidance trajectories. The guidance trajectories are optimized by the local planners in parallel, and the results are compared to determine which trajectory gets executed. T-MPC is faster, more consistent, and safer than several state-of-the-art planners. Contrary to similar existing work, it does not rely on an explicit lane structure and therefore enables both urban driving and mobile robotic applications.

The motion planners developed in this thesis are extensively validated in simulation and in experiments with a small-scale mobile robot and a full-scale self-driving vehicle navigating among pedestrians. The robot-agnostic implementation of the proposed planners that were developed for this thesis is available open source.

This thesis presents a comprehensive approach to integrating a trajectory planner and follower for autonomous vehicles (AVs) using model predictive contouring control (MPCC). The planner generates collision-free trajectories with a kinematic bicycle model, while the follower tracks them using a dynamic bicycle model with a smaller integration step size and higher update frequency. The hierarchical architecture allows for a long planning horizon and a fast control loop. Mismatches between the planner and follower can result in tracking errors and conservative trajectories. To address this, a feedback-hierarchical interface is proposed, feeding back the mismatch error from follower to planner. Obstacles are then inflated with this error, minimizing harmful deviations and collision risks. The paper validates the Local Motion Planner using a simulator with a dynamic bicycle model, testing different follower-solver settings in urban scenarios. The results show a reduction in collision rate of 23\% compared to a single-layer MPCC planner, with similar levels of lateral error and task duration.

Many autonomous navigation tasks require mobile robots to operate in dynamic environments involving interactions between agents. Developing interaction-aware motion planning algorithms that enable safe and intelligent interactions remains challenging. Dynamic game theory renders a powerful mathematical framework to model these interactions rigorously as coupled optimization problems. By solving the resultant coupled optimization problems to equilibrium solutions, the game-theoretic models explicitly account for the interdependence of agents’ decisions and achieve simultaneous prediction and planning. Coupled constraints between players, such as collision avoidance, can also be handled explicitly. However, most existing game-theoretic motion planning approaches rely on known objective models of all agents. This assumption presents a key obstacle to real-world ego-centric planning applications of these methods, where only local information is available. This thesis investigates solution approaches to relax this assumption and explicitly account for the ego agent’s uncertainty about other agents’ objectives while adaptively conducting game-theoretic motion planning.

The main contribution of this work is an online adaptive model-predictive game-play (MPGP) framework that jointly infers other players’ objectives and computes corresponding generalized Nash equilibrium (GNE) strategies. These strategies are then used as predictions for other players and control strategies for the ego agent. The adaptivity of the proposed approach is enabled by differentiating through a trajectory game solver whose gradient signal is used for maximum likelihood estimation (MLE) of opponents’ objectives. Compared with existing objective inference solutions in dynamic games, the proposed approach handles general inequality constraints in games and further supports direct integration with other differentiable modules, such as neural networks (NNs). Two simulation experiments indicate that the proposed approach performs closely to solving games with known objectives and outperforms the game-theoretic and model-predictive control (MPC) baselines. Two hardware experiments further demonstrate the real-time planning capability of the planner and its real-world applicability.

In addition to this main contribution, the second contribution of this work is a variational autoencoder (VAE) pipeline built upon the proposed differentiable game solver. This contribution aims at going beyond the point estimation in the first contribution and inferring potentially multi-modal beliefs about players’ objectives based on observations. The main idea is to employ variational inference (VI) to approximate Bayesian inference of players’ objectives. The variational autoencoder (VAE) framework is utilized for amortization to avoid per-sample optimization. Initial results on a single-player example show that after training, the proposed pipeline can: (i) generate a game objective distribution that resembles the underlying training data distribution and (ii) accurately predict a narrow, uni-modal posterior objective distribution when the observation is unambiguous based on seen data in the past and (iii) generate a multi-modal belief distribution of player’s objective to capture mostly likely modes in case of high uncertainty.

Creating autonomous Micro Aerial Vehicles for executing complex missions poses various challenges, including safe navigation in the presence of external wind disturbances. Most current navigation methods handle external wind disturbances through real-time estimation and rejection algorithms in the control stage, but lack safety guarantees in strong winds. Recent robust methods provide safety guarantees but can be overly conservative. With the availability of more powerful computing devices, data-driven control algorithms are becoming increasingly feasible. Combining Gaussian Process models with Model Predictive Control has shown to enhance safety and performance in various control applications. Moreover, Model Predictive Control has been extended to solve more complex optimization algorithms that combine trajectory generation and tracking, preventing reference trajectories that are risky and challenging to control in the presence of disturbances.

This research aims to improve quadrotor navigation in windy environments by using Gaussian Processes to model wind disturbances. The Gaussian Process model is integrated with a Model Predictive Contouring Control formulation that combines trajectory generation and control into one optimization problem. A nominal model of the quadrotor is derived and the Gaussian Process disturbance model is trained with the quadrotor position as input and wind disturbance as output. The wind disturbance map, along with the nominal model, is implemented in the Model Predictive Contouring Controller to consider both the mean and uncertainty of the resulting probabilistic model.

This study validates the use of Gaussian Processes to model complex wind disturbances in quadrotor navigation. The wind disturbance map is trained from available state information, with data collected using an optimal exploration design to minimize uncertainty and reduce exploration times. The hyperparameters involved in training the Gaussian Process model are discussed and the implementation of sparse Gaussian Processes is outlined to make it real-time feasible for the Model Predictive Contouring Control formulation. Including the prediction model by incorporating chance constraints results in improved tracking and increased robustness in cluttered environments compared to the nominal Model Predictive Contouring Control formulation. The proposed algorithm is shown to outperform state-of-the-art methods for safe quadrotor navigation in windy and cluttered environments, being able to handle more complex wind fields than existing methods while also being less conservative.

The ‘New Space’ mentality is gaining in popularity and is at the basis of the growing size of satellite constellations. These satellite constellations are used for technologies such as satellite navigation and internet, but a clear framework to controlling large satellite constellations is missing. Therefore, a new approach is developed to work efficiently with thousands of satellitesby finding a suitable model, controller and problem formulation.

A new model is developed that is linear time-invariant, fully based on physics and can include J2 perturbations. Furthermore, it can work well for both in-plane and out-of-plane movements. This model is paired with the lumped System-Level Synthesis control framework: a robust control algorithm that optimises the closed-loop transfer function. Due to a modification, the applied controller is less conservative than the original controller and faster than other modifications. The problem formulation is rewritten to a standard quadratic problem to significantly increase the rate at which these problems can be solved. This includes rewriting
one-norms and infinity norms, but also a new formulation for these System-Level Synthesis problems in general.

These findings are tested in a simulation of over two hundred satellites, where the satellites are controlled with the new model, the robust controller, the new problem formulation and collision avoidance constraint. These collision avoidance constraints are added to ensure that satellites keep a safe distance between themselves at all times. This includes constraints between satellites within the same plane, but also between satellites that cross each other’s orbit while close to each other.

Electric vehicles are a cleaner and more efficient means of transport. However, the sub-energy-optimal acceleration and deceleration inputs of drivers result in speed trajectories that cause superfluous expenditure of the stored electrical energy in battery. Optimising the speed trajectories to minimise the consumption of stored energy is a potential strategy for the efficient operation of electric vehicles. In this thesis, we propose a numerical solution to the eco-driving problem by optimising the speed trajectories via Pontryagin’s Minimum Principle. The solution is robust to the vehicle parameters and the driving conditions, and is used to generate energy-aware driving advice for near-straight line driving manoeuvres. In the end, we test the global optimality of the speed trajectories by convexification of the problem.

Cooperative driving controllers are becoming interesting subjects for future research in automated driving with the increase in connectivity. Using true-scale
autonomous vehicles to properly test new control algorithms can be a challenge
due to three main factors: costs, safety, and space requirements. To overcome
these problems, scaled-down vehicle platforms or so-called Remote Control (RC)
car platforms can be used to test algorithms in a safe environment without the
large costs involved with true-scale platforms. At present, existing RC platforms
are unnecessarily expensive or not able to drive autonomously.

This thesis presents the design of a low-cost 1/12 scale RC platform, consisting
of modified JetRacer Pro AI’s called R2C Car’s. The sensor suite consists of
a 160FoV 8MP camera and an added 2D LiDAR. These are used in a sensor
fusion and filtering framework to detect and locate other R2C Cars. Furthermore, wheel encoders are added to retrieve a velocity estimate of the R2C Car.
To showcase the possibilities of the R2C platform, a longitudinal cooperative
controller is implemented. This controller is based upon a Distributed Model
Predictive Control algorithm, whereby 5Ghz WiFi is used for the communication between R2C Cars. Evaluation of the R2C platform together with the
cooperative controller demonstrates that the platform is suitable for cooperative
driving experiments. Furthermore, information is presented on the performance
of the distributed algorithm.

Due to the continuously increasing volume of road freight transportation, there is a need to aid or automate truck-trailer driving. Truck docking, the process of parking a truck-trailer combination at a loading dock, is one of the most difficult manoeuvres for professional drivers. In this work, we propose a Model Predictive Control (MPC)-based truck docking driver assistance system. The objective of the system is to support the truck driver while parking the vehicle by means of visual instructions. MPC is an advanced control method that can be used to control Multiple-Input and Multiple-Output (MIMO) systems based on a finite horizon optimization. The control structure allows one to formulate a multi-objective control strategy with explicit constraints. This work has been conducted within the scope of the VIsion Supported Truck docking Assistant (VISTA) project and the practical application of the proposed system is to experiment with an MPC-based approach for the VISTA system currently in development. To determine the optimal control action, the proposed MPC-based driver assistant makes use of a kinematic model describing the motion of the vehicle as well as a second-order linear time-invariant model representing the driver’s behaviour. The system is examined by testing four performance categories: path tracking error, robustness, computational speed, and driver acceptance. The performance of the system is tested with professional truck drivers and people without a truck driver’s licence in a Virtual Reality (VR) simulation. The results suggest that the proposed MPC-based driver assistant is able to perform well regarding reference path tracking and is robust against driver errors, with satisfactory computational speed. The feedback and comments from the professional drivers during testing also indicate definite possible advantages of using the system. As of now, the MPC-based solution is preferred over previous concepts of the VISTA system.

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.

Mobile robots are getting more common in warehouses, distribution centers and factories, where they are used to boost productivity. At the same time, they are moving into the everyday world, where they need to operate in uncontrolled and cluttered environments. In order to extend the set of tasks that a robot can autonomously accomplish in such environments, we can equip mobile bases with a pushing skill. Possible use cases are delivering packages by pushing objects to a goal location, or clearing a path by pushing movable obstacles out of the way.

This thesis considers pushing with a non-holonomic mobile robot in cluttered environments. We develop a learning-based method, based on ensembles of probabilistic neural networks for learning the object’s dynamics. The models capture the inherent stochasticity of pushing from the variability of friction. In addition, the ensembles can capture the uncertainty that arises from a lack of data, so we can reason about which regions to avoid during the pushing and do not come into unrecoverable states.

The model is combined with a Model Predictive Path Integral (MPPI) controller. Each timestep the controller optimizes for a finite time horizon, which allows us to recover from slight model inaccuracies. The MPPI controller is capable of handling complex cost functions, which allows us to include the estimated uncertainty from the model to avoid uncertain regions. For obstacle-aware pushing, we provide the MPPI controller with a costmap grid, that contains the free pushing space, which the robot and object can use to maneuver to the goal location. Both the robot and object are constrained to keep within the free pushing space, which is constructed using LiDAR measurements.

We verify the method in simulation and compare it to two baseline methods: the state-of-the-art adaptive pushing controller proposed by Krivic et al. [31] and a pushing policy trained with the model-free reinforcement learning baseline Soft Actor-Critic (SAC). All three baselines reach comparable success rates in simulation of 100%, 98% and 96%, respectively. Furthermore, the learning-based MPPI shows improvements of 50% and 23% in accuracy compared to the baselines. The results are consistent for pushing in cluttered environments, where we show an increase of accuracy by 43%, compared to Krivic et al. [31].

Lastly, we validate the approach in a real-world setting, where the robot and object are tracked with a motion capture system. The approach has an overall success rate of 94% and performs successful pushing in a cluttered environment.