3D Gaussian splatting (3DGS) has recently demonstrated significant potential in computer vision, enabling high-fidelity 3D scene reconstruction with real-time rendering and fast training times. However, existing methods struggle in large, visually sparse, geometric self-similarity environments due to heavy reliance on image-based feature matching and depth information. In this work, we propose a novel reconstruction pipeline that reduces the dependence on visual features by incorporating IMU and LiDAR data to generate accurate point clouds and robustly localize images within the scene. Global colorization is achieved through 3D-to-2D projections of the localized images, which are then used to supervise 3DGS training. Our results demonstrate that the proposed pipeline significantly enhances the quality of 3D reconstruction for large, sparse scenarios, opening up new opportunities for applications in remote mapping and autonomous inspection.

Multi-agent path planning is a critical challenge in robotics, requiring agents to navigate complex environments while avoiding collisions and optimizing travel efficiency. This work addresses the limitations of existing approaches by combining Gaussian belief propagation with path integration and introducing a novel tracking factor to ensure strict adherence to global paths. The proposed method is tested with two different global path-planning approaches: rapidly exploring random trees and a structured planner, which leverages predefined lane structures to improve coordination. A simulation environment was developed to validate the proposed method across diverse scenarios, each posing unique challenges in navigation and communication. Simulation results demonstrate that the tracking factor reduces path deviation by 28% in single-agent and 16% in multi-agent scenarios, highlighting its effectiveness in improving multi-agent coordination, especially when combined with structured global planning.

Physics-informed neural networks (PINNs) integrate physical laws with data-driven models to improve generalization and sample efficiency. This work introduces an open-source implementation of the Physics-Informed Neural Network with Control (PINC) framework, designed to model the dynamics of an underwater vehicle. Using initial states, control actions, and time inputs, PINC extends PINNs to enable physically consistent transitions beyond the training domain. Various PINC configurations are tested, including differing loss functions, gradient-weighting schemes, and hyperparameters. Validation on a simulated underwater vehicle demonstrates more accurate long-horizon predictions compared to a non-physics-informed baseline.

In autonomous and mobile robotics, a principal challenge is resilient real-time environmental perception, particularly in situations characterized by unknown and dynamic elements, as exemplified in the context of autonomous drone racing. This study introduces a perception technique for detecting drone racing gates under illumination variations, which is common during high-speed drone flights. The proposed technique relies upon a lightweight neural network backbone augmented with capabilities for continual learning. The envisaged approach amalgamates predictions of the gates' positional coordinates, distance, and orientation, encapsulating them into a cohesive pose tuple. A comprehensive number of tests serve to underscore the efficacy of this approach in confronting diverse and challenging scenarios, specifically those involving variable lighting conditions. The proposed methodology exhibits notable robustness in the face of illumination variations, thereby substantiating its effectiveness.

This work presents a novel approach which integrates deep learning, online learning and continual learning paradigms for adaptive control for robotic systems. Deep learning allows generalising knowledge about the robot, while online learning can adapt to variable operating conditions, and continual learning enables remembering previous knowledge. The proposed method approximates the inverse dynamics of the robot, which is formulated as a regression problem. With a minimum knowledge of the robot's dynamics, the proposed method shows its capability to reduce tracking errors online by continuously learning and compensating for internal and external changing conditions. Furthermore, the simulation results show that the proposed approach with online continual learning improves the control performance of ground and aerial mobile robots.

Nowadays, wheeled mobile robots constitute a considerable portion of robots in industrial applications. Generally, regardless of their purpose, these systems are not designed to physically interact with humans, other robots, or the environment. In this study, we present a novel safe autonomous mobile - SAM - robot, which is a torque-controlled compliant robot that is conceived for safe human-robot interaction. This work provides an overview of the development philosophy of the system, its mechanical and mechatronics structure along with control and navigation architecture. Preliminary results show the advantages of the proposed mobile robot while interacting with its surroundings. We believe that this study will bring the wheeled mobile robots one step closer to the proactive interaction with their environment and humans surrounding them.

We present an accessible robot interface for telemanipulation (A-RIFT), which preserves the haptic channel partially in a zero-additional-cost interface by visual substitution of force feedback (VSFF). This work explores a gap in the literature, resulting from the focus on performance improvements in telerobotics at increasing interface costs. Unlike most telemanipulation interfaces for high-degree-of-freedom robotic systems, this one requires minimal training and can be run in a web browser under high latency conditions, using an Internet connected computer with the user's own mouse and keyboard. To evaluate the performance of the system, we ran a controlled user study (N=12) to test how different distances (local vs. remote) and VSFF (on vs. off) affect the system's usability. As expected, participants in remote conditions performed worse than those in closer proximity. Despite several participants claiming that the visual display of force feedback did not help them, our analysis of their task performance showed that operators in remote condition actually performed statistically significantly better with the visual force feedback display than without it. These results indicate a promising new interface design direction for low-cost telemanipulation.

Fast and robust gate perception is of great importance in autonomous drone racing. We propose a convolutional neural network-based gate detector (GateNet1) that concurrently detects gate’s center, distance, and orientation with respect to the drone using only images from a single fish-eye RGB camera. GateNet achieves a high inference rate (up to 60 Hz) on an onboard processor (Jetson TX2). Moreover, GateNet is robust to gate pose changes and background disturbances. The proposed perception pipeline leverages a fish-eye lens with a wide field-of-view and thus can detect multiple gates in close range, allowing a longer planning horizon even in tight environments. For benchmarking, we propose a comprehensive dataset (AU-DR) that focuses on gate perception. Throughout the experiments, GateNet shows its superiority when compared to similar methods while being efficient for onboard computers in autonomous drone racing. The effectiveness of the proposed framework is tested on a fully-autonomous drone that flies on previously-unknown track with tight turns and varying gate positions and orientations in each lap.

Drone racing is a recreational sport in which the goal is to pass through a sequence of gates in a minimum amount of time, while avoiding collisions. In autonomous drone racing, one must accomplish this task by flying fully autonomously in an unknown environment by relying only on computer vision methods for detecting the target gates. Due to the challenges such as background objects and varying lighting conditions, traditional object detection algorithms based on colour or geometry tend to fail. Convolutional neural networks offer impressive advances in computer vision, but require an immense amount of data to learn. Collecting this data is a tedious process because the drone has to be flown manually, and the data collected can suffer from sensor failures. In this work, a semi-synthetic dataset generation method is proposed, using a combination of real background images and randomised 3D renders of the gates, to provide a limitless amount of training samples that do not suffer from those drawbacks. Using the detection results, a line-of-sight guidance algorithm is used to cross the gates. In several experimental real-time tests, the proposed framework successfully demonstrates fast and reliable detection and navigation.

This article presents an online learning method for improved control of nonlinear systems by combining deep learning and fuzzy logic. Given the ability of deep learning to generalize knowledge from training samples, the proposed method requires minimum amount of information about the system to be controlled. However, in robotics, particularly in aerial robotics where the operating conditions may vary, online learning is required. In this article, fuzzy logic is preferred to provide supervising feedback to the deep model for adapting to variations in the system dynamics as well as new operational conditions. The learning method is divided into two phases: offline pretraining and online posttraining. In the former, the system is controlled by a conventional controller and a deep fuzzy neural network (DFNN) is pretrained based on the recorded input-output dataset, in order to approximate the inverse dynamical model of the system. In the latter, only the pretrained DFNN is used to control the system. In this phase, the fuzzy logic, which encodes the expert knowledge, is utilized to observe the behavior of the system and to correct the action of DFNN instantaneously. The experimental results show that the proposed online learning-based approach improves the trajectory tracking performance of the unmanned aerial vehicle.

This work presents an online learning-based control method for improved trajectory tracking of unmanned aerial vehicles using both deep learning and expert knowledge. The proposed method does not require the exact model of the system to be controlled, and it is robust against variations in system dynamics as well as operational uncertainties. The learning is divided into two phases: offline (pre-)training and online (post-)training. In the former, a conventional controller performs a set of trajectories and, based on the input-output dataset, the deep neural network (DNN)-based controller is trained. In the latter, the trained DNN, which mimics the conventional controller, controls the system. Unlike the existing papers in the literature, the network is still being trained for different sets of trajectories which are not used in the training phase of DNN. Thanks to the rule-base, which contains the expert knowledge, the proposed framework learns the system dynamics and operational uncertainties in real-time. The experimental results show that the proposed online learning-based approach gives better trajectory tracking performance when compared to the only offline trained network.

As non-singleton fuzzy logic controllers (NSFLCs) are capable of capturing input uncertainties, they have been effectively used to control and navigate unmanned aerial vehicles (UAVs) recently. To further enhance the capability to handle the input uncertainty for the UAV applications, a novel NSFLC with the recently introduced similarity-based inference engine, i.e., Sim-NSFLC, is developed. In this paper, a comparative study in a 3D trajectory tracking application has been carried out using the aforementioned Sim-NSFLC and the NSFLCs with the standard as well as centroid composition-based inference engines, i.e., Sta-NSFLC and Cen-NSFLC. All the NSFLCs are developed within the robot operating system (ROS) using the C++ programming language. Extensive ROS Gazebo simulation-based experiments show that the Sim-NSFLCs can achieve better control performance for the UAVs in comparison with the Sta-NSFLCs and Cen-NSFLCs under different input noise levels.

A significant number of investigations of type-1 and type-2 fuzzy logic controllers have revealed their exceptional ability to capture uncertainties in complex and nonlinear systems, particularly in real-time control applications. However, regardless of being type-1 or type-2, fuzzy logic controller design is still a complicated task due to the lack of a closed form solution of the output and an interpretable relationship between the control output and fuzzy logic controller design parameters, such as center or width of the membership functions. To simplify the design procedure further, we think every attempt to obtain such interpretable relationships is worthwhile. Accordingly, this paper aims to design a double-input interval type-2 fuzzy PID controller and obtain interpretable relationships between the input and the output of the controller. Thereafter, we deploy the novel design for the control of a Y6 coaxial tricopter unmanned aerial vehicle. Simulation results, which are realised in robot operating system (ROS) using C++ and Gazebo environment, are found to tally with the theoretical analysis and claims in the paper.

Fuzzy logic controllers (FLCs) have extensively been used for the autonomous control and guidance of unmanned aerial vehicles (UAVs) due to their capability of handling uncertainties and delivering adequate control without the need for a precise, mathematical system model which is often either unavailable or highly costly to develop. Despite the fact that non-singleton FLCs (NSFLCs) have shown more promising performance in several applications when compared to their singleton counterparts (SFLCs), most of UAV applications are still realized by using SFLCs. In this paper, we explore the potential of both standard and the recently introduced centroid based NSFLCs, i.e., Sta-NSFLC and Cen-NSFLC, for the control of a quadcopter UAV under various input noise conditions using different levels of fuzzifier, and a comparative study has been conducted using the three aforementioned FLCs. We present a series of simulation-based experiments, the simulation results show that the control performances of NSFLCs are better than those of SFLC, and the Cen-NSFLC outperforms the Sta-NSFLC especially under highly noisy conditions.