Projects

a collection of my most recent works

Curated below is a selection of projects that encapsulate my primary research interests and standout achievements. Most of these projects were developed in collaboration with other students and colleagues in multidisciplinary teams. For access to project code and additional references, kindly reach out to me by email. Your interest and inquiries are greatly appreciated.

A model predictive control approach for trajectory tracking of quadrotors

Our paper presents a model predictive control (MPC) approach for trajectory tracking of quadrotors, where a mathematical model of the quadrotor is developed with consideration of aerodynamic moment and drag to improve the simulation’s real-world accuracy.

Autonomous legged robot solution for efficient lost item retrieval

One of the specific investigations of TNO concerns the applicability of Boston Dynamics’ Spot robot for domestic applications, i.e., applying SPOT as a robotic assistance dog.

Design of a drone swarm to measure the wind field and temperature stratification in a wind farm

The goal of this project is to outline the sub-system design of the local sensing system, to satisfy the mission need statement: measure the atmospheric conditions with full three-dimensional coverage of a wind farm to optimize its operational performance and control.

Developing analytical models for dynamic analysis of lattice structures

Lattice structures have become ever more popular in modern structural design, as they generally provide high strength-to-weight ratios. However, certain designs also prove to be effective in suppressing waves of specific frequencies, allowing lattice materials to be used as for instance noise isolators.

Enhancing autonomous driving: leveraging frame‐based foundation models for improved pedestrian detection and sensor fusion

Foundational models learn and generalize from large-scale image datasets created by scraping the entire web, meaning driving-related images (e.g. dashcam footage) could be part of the model’s knowledge. With current advancements in foundational computer vision models, they could improve sensor fusion pipeline by increasing initial pedestrian detection accuracies and better complement the hand-crafted sensor fusion step.

Obstacle driven RRT implementation for quad‐rotor motion planning

This paper presents the development of a version of RRT* optimized for indoor environments, called biased RRT*.

Reproduction of deep learning paper on event camera processing

In this paper, we reproduce the AEGNN model and evaluate its performance in recognizing objects in the Neuromorphic-Caltech101 dataset. We investigate the impact of changing convolutional operators and hyperparameters on the model’s accuracy.

Robotic particle image velocimetry for position reconstruction of a smart rotor model

Smart Rotor Models (SRM) allow for modification of the wind turbine’s aerodynamic properties using an actuated flap to alleviate the effects of sudden peak loads caused by wind gusts. Wind tunnel experiments conducted with the help of Robotic Particle Image Velocimetry (RPIV) measure the flow surrounding such models.