document

Learning Vision-based Navigation Policies for Information Gathering with Quadcopters: A Deep Reinforcement Learning Approach

Vellekoop, Joris (author)

Deep reinforcement learning presents a compelling approach for the exploration of cluttered 3D environments, offering a balance between fast computation and effective vision-based navigation. Yet, the use of 3D navigation for learning-based information gathering remains largely unexplored. Navigation in 3D space poses the challenge of having an...

master thesis 2024

document

Towards a Better Understanding of Attitude Oscillations Experienced by Quadcopters In Axial Descent

Bogaerts, Charlie (author)

This study aims to quantify the aerodynamic behaviour of fixed pitch rotors with a diameter of 12.7 cm in axial descent, and its contribution to the attitude oscillations found for quadcopters under similar flight conditions. Wind tunnel tests are performed with an isolated rotor, as well as a complete quadcopter mounted inside a gimbal that...

master thesis 2023

document

Safe Flight Envelope Prediction and Experimental Validation of Multirotor UAV: An Analysis on the Effects of Longitudinal Center of Gravity Position and Actuator Dynamics on Quadcopter Flight Envelopes

Kaffa, Lauren (author)

Loss of control (LOC) is the primary cause of failure of Unmanned Aerial Vehicles (UAV). The safety of these systems can be largely improved by facilitating techniques to prevent LOC to occur, such as Flight Envelope Protection, enabling controllers to keep the system within the Safe Flight Envelope (SFE).<br/>The aim of this work is to examine...

master thesis 2023

document

A Differential Flatness-based Trajectory Planner for Endpoint Control

van der Velde, Geert (author)

For increasing the safety of quadcopters the development of recovery control algorithms is crucial. A common cause of quadcopter crashes is collisions. To validate recovery control algorithms on collisions, the quadcopter has to reach the post-collision state. <br/>For this, the principle of 'endpoint control' is introduced, bringing the...

master thesis 2023

document

Estimating wind fields using drones in a network: Estimating hyperlocal wind fields with on-board sensors on quadcopters

van Baasbank, Erik (author)

Charting hyperlocal wind using a drone is a challenge of increased attention as it unlocks potential in a variety of fields. In context of the METeo Sensors In the Sky project, this study proposes a method to estimate the magnitude and direction of wind using a quadcopter in hover and cruise without a dedicated wind sensor. Only on-board sensors...

master thesis 2023

document

Improving the PX4 software-in-the-loop multirotor UAV simulator accuracy

te Braake, Michiel (author)

Recent developments have enabled the mass production of cheap, high-performance multirotors. As a result of this, the multirotor has found a large variety of different uses. Testing new algorithms using real-life flights is costly in terms of time and potentially in terms of materials in the case of a crash. Simulators have quickly gained...

master thesis 2022

document

Quadcopter Loss-of-Control Prediction Using Recurrent Neural Networks

Altena, Anique (author)

Loss-of-control (LOC) is the main cause of crashes for drones. On-board prevention systems should be designed that require low computing power and memory. Data-driven techniques serve as a solution. This study proposes the use of recurrent neural networks (RNN) for LOC prediction. The aim is to identify which RNN model is most suitable and if...

master thesis 2022

document

Reinforcement Learning for Quadcopter Fault Tolerance: Analyzing Hierarchical and Online Deep Reinforcement Learning for Quadcopter Fault Tolerant Control

Bello, Riccardo (author)

The demand of adding fault tolerance to quadcopter control systems has significantly increased with the rise of adoption of UAVs in numerous sectors. This work proposes and demonstrates the use of Hierarchical Reinforcement Learning to control a quadcopter subject to severe actuator fault. State-of-the-art algorithms are implemented, and a...

master thesis 2021

document

Improving the Performance of INDI Flight Control for a Quadrotor in the Ceiling Effect

Kemmeren, Max (author)

As the application areas of Unmanned Aerial Vehicles (UAVs) keep expanding, new flight areas are encountered more often. Small UAVs, named Micro Air Vehicles (MAVs), even fly in areas like sewage pipes. These areas introduce new difficulties such as aerodynamic effects caused by the ground and/or ceiling. In this paper two main contributions are...

master thesis 2021

document

Attitude Estimation of a Quadcopter with one fully damaged rotor using on-board MARG Sensors

Solanki, Prashant (author)

Quadcopters are becoming increasingly popular across diverse sectors such as mapping, photography, or surveillance. Since rotor damages occur frequently, it is essential to improve the attitude estimation and thus ultimately the ability to control a damaged quadcopter. The Control and Simulation group of TU Delft developed a quadcopter...

master thesis 2020

document

Sample-efficient reinforcement learning for quadcopter flight control

Koomen, Lenard (author)

The combination of reinforcement learning and deep neural networks has the potential to train intelligent autonomous agents on high dimensional sensory inputs, with applications in flight control. However, the amount of samples needed by these methods is often too large to use real-world interaction. In this work, mirror-descent guided policy...

master thesis 2020

document

Low level quadcopter control using Reinforcement Learning: Developing a self-learning drone

Koning, Tim (author)

Reinforcement Learning (RL) is a learning paradigm where an agent learns a task by trial and error. The agent needs to explore its environment and by simultaneously receiving rewards it learns what is appropriate behaviour.<br/>Even though it has roots in machine learning, RL is essentially different from other machine learning methods. In...

master thesis 2020

document

The Landing of a Quadcopter on Inclined Surfaces using Reinforcement Learning

Fris, Rein (author)

Deep Reinforcement Learning (DRL) enables us to design controllers for complex tasks with a deep learning approach. It allows us to design controllers that are otherwise cumbersome to design with conventional control methodologies. Often, an objective for RL is binary in nature. However, exploring in environments with sparse rewards is a problem...

master thesis 2020

document

Network Decentralized Collision Avoidance with Applications in a Scalable Unmanned Aerial System Testbed

Ledzian, Patrick (author)

Decentralized control and estimation are both active research areas in the field of systems and control. A new approach to these topics utilizes graph theory to characterize inter-agent communication as a graph that, in this thesis, can have time-varying topology. This approach has been named "network-decentralized" and the use of network...

master thesis 2019

document

Obstacle avoidance for quadrotors using reinforcement learning and obstacle-airflow interactions

van Dam, Geart (author)

This research investigates and proposes a new method for obstacle detection and avoidance on quadrotors. One that does not require the addition of any sensors, but relies solely on measurements from the accelerometer and rotor controllers. The detection of obstacles is based on the principle that the airflow around a quadrotor changes when the...

master thesis 2019

document

Flight control and collision avoidance for quadcopter and flapping wing MAVs using only optical flow: Theory, Simulation and Experiment

van Vrede, Daan (author)

Both quadcopter Micro Aerial Vehicles (MAVs) and Flapping Wing MAVs (FWMAVs) are constrained in Size, Weight and Processing power (SWaP) in order to achieve flight tasks that include attitude and velocity stabilisation, as well as obstacle avoidance. <br/>Conventional sensory and control approaches, such as those relying on inertial, visual and...

master thesis 2018

document

Vision-based stabilization of micro quadrotors

Braber, T.I. (author)

On-board stabilization of quadrotors is often done using an Inertial Measurement Unit (IMU), aided by additional sensors to combat the IMU drift. For example, GPS readings can aid when flying outdoors, or when flying in GPS denied environments, such as indoors, visual information from one or more camera modules can be used. <br/>A single...

master thesis 2017

document

Continuous state and action Q-learning framework applied to quadrotor UAV control

Naruta, Anton (author)

This paper describes an implementation of a reinforcement learning-based framework applied to the control of a multi-copter rotorcraft. The controller is based on continuous state and action Q-learning. The policy is stored using a radial basis function neural network. Distance-based neuron activation is used to optimize the generalization...

master thesis 2017

document

Investigation of L1 Adaptive Control as an Augmentation Loop

Vicente Melo E Carvalho Marques, Mariana (author)

L1 adaptive control is a relatively new technique that attempts to tackle the robustness shortcoming of the MRAC controller by applying a low pass filter to the control input. It can be used as an augmentation loop using baseline controllers such as backstepping or PID. This controller has been implemented as an augmentation loop with both these...

master thesis 2017

document

Differential Dynamic Programming for Aerial Robots using an Aerodynamics Model

Abuter Grebe, Nicolás Omar (author)

State of the art trajectory generation schemes for quadrotors assume a simple dynamic model. They neglect aerodynamic effects such as induced drag and blade flapping and assume that no wind is present. In order to overcome this limitation, this thesis investigates a trajectory optimization scheme based upon Differential Dynamic Programming (DDP)...

master thesis 2017