Doctoral Networks (DNs) aim to address systemic challenges in doctoral education, such as fostering interdisciplinarity, enabling international and intersectoral collaboration, enhancing employability, and promoting responsible innovation. While cohort-based training helps mitigate student isolation through workshops and summer schools, traditional DNs often struggle to fully realise their collaborative potential, often relying on predefined supervisor relationships or the initiative of individual researchers. In contrast, Marie Skłodowska-Curie Doctoral Networks (MSCA-DNs) prioritise doctoral candidates (DCs), challenging them to balance independent research with contributions to a shared, mission-driven objective. This study examines how structured training, including digital communities and application-focused research sprints, enhances system integration and collaboration within the Robotics and AI for Critical Asset Monitoring (RAICAM) Doctoral Network. DCs located across seven European countries worked in virtual teams, refining systems through structured workflows, weekly meetings, and shared workspaces before training schools. Through continuous online collaboration and targeted sprints, RAICAM facilitated interdisciplinary integration. Two research sprints, conducted in Italy and France, allowed teams to develop and test solutions for real-world challenges with an impact-driven plan that considers a given problem from and end-to-end perspective that requires and foster interdisciplinary collaboration. The results highlight the effectiveness of structured training in enhancing collaboration and adaptability, while identifying key areas for improvement. This study translates lessons from RAICAM into practical guidelines for future doctoral networks, demonstrating how structured training empowers students to drive interdisciplinary research independently.

As legged robots continue to evolve, new control methods are being developed to provide fast, robust, accurate and computationally efficient algorithms for traversing challenging environments. This paper presents a realtime adaptive locomotion controller for quadrupeds, designed to maintain stability and controllability on various surfaces, including highly slippery terrains. The proposed approach optimizes control effort distribution based on the probability of slippage by utilizing a surface-independent adaptation layer. By balancing the robot's redundant kinematic system through rank relaxation-similar to loosening constraints in optimization problems-this method demonstrates significant performance improvements. Unlike Reinforcement Learning (RL) approaches, which depend on pre-trained policies and may struggle to adapt velocity tracking control across different terrains, our method rapidly adjusts to changing conditions, as validated by extensive simulation experiments.

In this work, a control scheme for human-robot collaborative object transportation is proposed, considering a quadruped robot equipped with the MIGHTY suction cup that serves both as a gripper for holding the object and a force/torque sensor. The proposed control scheme is based on the notion of admittance control, and incorporates a variable damping term aiming towards increasing the controllability of the human and, at the same time, decreasing her/his effort. Furthermore, to ensure that the object is not detached from the suction cup during the collaboration, an additional control signal is proposed, which is based on a barrier artificial potential. The proposed control scheme is proven to be passive and its performance is demonstrated through experimental evaluations conducted using the Unitree Go1 robot equipped with the MIGHTY suction cup.

Anomaly detection is a fundamental task for time-series analytics with important implications for the downstream performance of many applications. Despite increasing academic interest and the large number of methods proposed in the literature, recent benchmark and evaluation studies demonstrated that there exists no single best anomaly detection method when applied to heterogeneous time series datasets. Therefore, the only scalable and viable solution to solve anomaly detection over very different time series collected from diverse domains is to propose a model selection method that will choose, based on time series characteristics, the best anomaly detection method to run. This paper describes ADecimo, a modular and extensible web application that helps users understand the performance of time series classification algorithms used as model selection methods for time series anomaly detection. Overall, our system enables users to compare 17 different classifiers over 1980 time series, and decide on the most suitable time series classification method for their own time series and use cases.

This poster introduces a novel bi-directional randomized sampling-based path planning algorithm coined as Ray Casting and Diffusion (RCD). RCD models both the robot and the target as light point-sources that emit pseudo-rays towards random directions. Upon collision with obstacles or same source produced rays (robot or target), a new source point is established and a weight coefficient is assigned to it that is used in subsequent iterations. The primary concept of RCD is to iteratively repeat this process until an intersection between the robot-generated-rays and the target-generated-rays occurs (path found). RCD incorporates multiple optimizations within each layer, contributing to the proposed approach's speed, reliability, probabilistic completeness, and its ability to consistently generate feasible and efficient paths. Both the implementation and the experimental datasets have been released as an open-source project to support future research endeavors. Extensive evaluations of RCD have been conducted, comparing it against multiple well-established state-of-the-art sampling-based path planning approaches. The results indicate superior efficiency in finding paths faster compared to existing methods. In some instances, RCD yields slightly longer paths, however its overall performance showcases significant advantages in terms of speed and efficacy in path finding tasks.

Nowadays, robotic arms are ubiquitously employed for object manipulation across a spectrum of applications, spanning from production lines to warehouses, and encompassing both stationary and mobile robotic systems. Among the most prevalent end-effectors, used for the majority of these applications, are suction cups. The rudimentary act of grasping an object and relocating it, devoid of a cognizant awareness of the forces stemming from the object’s motion and grip, can result in suboptimal and inefficient robot movements. In more dire circumstances, such negligent handling may precipitate detachment of the object from the end-effector, potentially incurring damage to either the object or the arm.In this paper, we build upon the advanced sensing and attaching capabilities of our suction cup MIGHTY, and introduce GEM-C, a novel Gravity, External forces and Motion Compensation controller, that constantly adapts the orientation of the suction cup so as to enhance the quality of attachment. Throughout all examined scenarios and experiments, our approach remarkably improved the robot’s performance by providing the optimal end-effector pose while also reducing the stress on the motors and the overall power consumption. The derived results, clearly demonstrate the MIGHTY and GEM-C schema’s potential for a wide range of demanding robotic manipulation tasks.

Anomaly detection is a fundamental task for time-series analytics with important implications for the downstream performance of many applications. Despite increasing academic interest and the large number of methods proposed in the literature, recent benchmark and evaluation studies demonstrated that no overall best anomaly detection methods exist when applied to very heterogeneous time series datasets. Therefore, the only scalable and viable solution to solve anomaly detection over very different time series collected from diverse domains is to propose a model selection method that will select, based on time series characteristics, the best anomaly detection method to run. Existing AutoML solutions are, unfortunately, not directly applicable to time series anomaly detection, and no evaluation of time series-based approaches for model selection exists. Towards that direction, this paper studies the performance of time series classification methods used as model selection for anomaly detection. Overall, we compare 17 different classifiers over 1800 time series, and we propose the first extensive experimental evaluation of time series classification as model selection for anomaly detection. Our results demonstrate that model selection methods outperform every single anomaly detection method while being in the same order of magnitude regarding execution time. This evaluation is the first step to demonstrate the accuracy and efficiency of time series classification algorithms for anomaly detection, and represents a strong baseline that can then be used to guide the model selection step in general AutoML pipelines.

The spectrum of applications of robotic systems is constantly being expanded in the research, industrial and even the defense sectors, ranging from manipulation and assembly to critical infrastructure monitoring and post-disaster response. Nevertheless, contemporary robotic capabilities are significantly hindered when traversing through or interacting with complex, unstructured and dynamic environments. To alleviate for that, we introduce in the current work MIGHTY (Multi-functional Intelligent Gripping with High Tolerance), a novel, lightweight, sensor-enhanced vacuum suction cup providing not only enhanced attachment capabilities on a plethora of surfaces of varying roughness, but also robust and accurate contact and force estimation. Its unique design facilitates multiple functionalities, from acting as a gripper for object manipulation to operating as a foot for stable walking and steep surface climbing. The proposed suction cup was extensively assessed under varying experimental setups, in order to validate its capacity to sense the applied force and torque and the torque's axis, as well as its ability to attach on a variety of surfaces. In all cases remarkable results were demonstrated, attesting for its effectiveness and robustness.

Visual Simultaneous Localization and Mapping (vSLAM) is the method of employing an optical sensor to map the robot’s observable surroundings while also identifying the robot’s pose in relation to that map. The accuracy and speed of vSLAM calculations can have a very significant impact on the performance and effectiveness of subsequent tasks that need to be executed by the robot, making it a key building component for current robotic designs. The application of vSLAM in the area of humanoid robotics is particularly difficult due to the robot’s unsteady locomotion. This paper introduces a pose graph optimization module based on RGB (ORB) features, as an extension of the KinectFusion pipeline (a well-known vSLAM algorithm), to assist in recovering the robot’s stance during unstable gait patterns when the KinectFusion tracking system fails. We develop and test a wide range of embedded MPSoC FPGA designs, and we investigate numerous architectural improvements, both precise and approximation, to study their impact on performance and accuracy. Extensive design space exploration reveals that properly designed approximations, which exploit domain knowledge and efficient management of CPU and FPGA fabric resources, enable real-time vSLAM at more than 30 fps in humanoid robots with high energy-efficiency and without compromising robot tracking and map construction. This is the first FPGA design to achieve robust, real-time dense SLAM operation targeting specifically humanoid robots. An open source release of our implementations and data can be found in [ 1 ].

During pressure support ventilation (PSV), excessive assist results in weak inspiratory efforts and promotes diaphragm atrophy and delayed weaning. The aim of this study was to develop a classifier using a neural network to identify weak inspiratory efforts during PSV, based on the ventilator waveforms. Recordings of flow, airway, esophageal and gastric pressures from critically ill patients were used to create an annotated dataset, using data from 37 patients at 2-5 different levels of support, computing the inspiratory time and effort for every breath. The complete dataset was randomly split, and data from 22 patients (45,650 breaths) were used to develop the model. Using a One-Dimensional Convolutional Neural Network, a predictive model was developed to characterize the inspiratory effort of each breath as weak or not, using a threshold of 50 cmH(2)O*s/min. The following results were produced by implementing the model on data from 15 different patients (31,343 breaths). The model predicted weak inspiratory efforts with a sensitivity of 88%, specificity of 72%, positive predictive value of 40%, and negative predictive value of 96%. These results provide a 'proof-of-concept' for the ability of such a neural-network based predictive model to facilitate the implementation of personalized assisted ventilation.

Quadruped robots are commonly employed in challenging and unpredictable environments where their stability and controllability could be compromised during locomotion. During task execution, the robot might face non-controllable conditions or loss of balance either due to dynamic contact events or aggressive control efforts, e.g. during trajectory tracking. In this work, we propose a robust adaptive trajectory tracking controller that consists of two prioritized layers of adaptation aimed at maintaining stability during dynamic (possibly unstable) contact of one or more supporting legs. The aforementioned key-feature can serve as the basis for avoiding slippage during dynamic locomotion. The proposed method is a novel lightweight analytical solution that assumes no prior knowledge of the friction properties of the surface. This is accomplished by considering the slippage probability as extracted by our previous work on contact state estimation in order to avoid non-controllable conditions. We evaluate our approach through simulations and validate it with real experiments, demonstrating robustness in multiple scenarios.

Legged robot navigation in unstructured and slippery terrains depends heavily on the ability to accurately identify the quality of contact between the robot's feet and the ground. Contact state estimation is regarded as a challenging problem and is typically addressed by exploiting force measurements, joint encoders and/or robot kinematics and dynamics. In contrast to most state of the art approaches, the current work introduces a novel probabilistic method for estimating the contact state based solely on proprioceptive sensing, as it is readily available by Inertial Measurement Units (IMUs) mounted on the robot's end effectors. Capitalizing on the uncertainty of IMU measurements, our method estimates the probability of stable contact. This is accomplished by approximating the multimodal probability density function over a batch of data points for each axis of the IMU with Kernel Density Estimation. The proposed method has been extensively assessed against both real and simulated scenarios on bipedal and quadrupedal robotic platforms such as ATLAS, TALOS and Unitree's GO1.

We propose a novel, end-to-end, decision support system, to facilitate ICU clinicians to identify ventilatory over-assistance and titrate the level of respiratory support. Our method consists of a wavelet-based algorithm to automatically segment distinct breaths in mechanical ventilator respiratory recordings, and a 1D CNN schema for the classification of new respirations. A dataset of 40 respiratory recordings, taken from 38 ICU patients, was used for quantitative performance assessment, where our approach achieved impressive results in detecting ventilatory over-assistance in a total of 76,595 distinct breath patterns. The proposed system is non-invasive, requires no changes in clinical practice and is readily applicable to contemporary mechanical ventilators. Accordingly, it facilitates efficient ventilator exploitation and reduces the risk of mechanical ventilation complications.

In this article, we propose a deep learning framework that provides a unified approach to the problem of leg contact detection in humanoid robot walking gaits. Our formulation accomplishes to accurately and robustly estimate the contact state probability for each leg (i.e., stable or slip/no contact). The proposed framework employs solely proprioceptive sensing and although it relies on simulated ground-truth contact data for the classification process, we demonstrate that it generalizes across varying friction surfaces and different legged robotic platforms and, at the same time, is readily transferred from simulation to practice. The framework is quantitatively and qualitatively assessed in simulation via the use of ground-truth contact data and is contrasted against state of-the-art methods with an ATLAS, a NAO, and a TALOS humanoid robot. Furthermore, its efficacy is demonstrated in base estimation with a real TALOS humanoid. To reinforce further research endeavors, our implementation is offered as an open-source ROS/Python package, coined Legged Contact Detection (LCD).

In this work we present a novel Supervised Learning scheme for executing sensitive Force-based manipulation tasks. The proposed scheme, termed SLF, is formulated as a three-stage process: (a) supervised trial-execution in simulation to acquire sufficient training data; (b) training to facilitate grasp learning with suitable robot-arm pose and lifting force; (c) grasp execution in simulation. Consequently, following sim-to-real transfer, operation in real environments is achieved in addition to simulated ones, generalizing also for objects not included in the trial sessions. The proposed learning scheme is demonstrated in object lifting tasks where the applied force varies for different objects with similar contact friction coefficients, and likewise the grasping pose. Experimental results on the manipulator YuMi show that the robot is able to effectively reproduce demanding lifting and manipulation tasks after learning is accomplished.

Visual Simultaneous Localization and Mapping (vSLAM) is the process of mapping the robot's observed environment using an optical sensor, while concurrently determining the robot's pose with respect to that map. For humanoid robots, the implementation of vSLAM is particularly challenging, due to the intricate motions of the robot. In this work, we present a pose graph optimization module based on RGB features, as an extension on the KinectFusion pipeline (a well-known vSLAM algorithm), to help recover the robot's pose during unstable gait patterns where the KinectFusion tracking system fails. We implement and evaluate a plethora of embedded MPSoC FPGA designs and we explore several architectural optimizations, both precise and approximate, highlighting their effect on performance and accuracy. Properly designed approximations, which exploit domain knowledge and efficient management of CPU and FPGA fabric resources, enable real-time vSLAM (at more than 30 fps) in humanoid robots without compromising robot tracking and map construction. We show that a combination of precise and approximate optimizations and tuning of algorithmic parameters provide a speedup of up to 15.7X and 22.5X compared with the precise FPGA and ARM-only implementations, respectively, without violating the tight accuracy constraints.

In the next few years, new demanding applications will be supported on mobile platforms by reconciling two conflicting requirements: high performance (often with real-time limitations) and low power consumption. The objective of the vipGPU project is to develop hardware and software technology to provide efficient support for two such application scenarios, namely (a) simultaneous localization and mapping (SLAM) in mobile robotics systems, and (b) virtual reality (VR) in portable devices to simulate serious games with emphasis on simulating surgical interventions and medical training in general. In this project, we aim at developing a new heterogeneous platform consisting of hardware accelerators for low power embedded systems optimized (at the hardware and software level) for the implementation of the two applications mentioned above.

In the current paper we investigate the challenges of localizing walking humanoid robots using Visual SLAM (VSLAM). We propose a novel dense RGB-D SLAM framework that seamlessly integrates with the dynamic state of a humanoid, to provide real-time localization and dense mapping of its surroundings. Following the path of recent research in humanoid localization, in the current work we explore the integration between a VSLAM system and the humanoid state, by considering the gait cycle and the feet contacts. We analyze how these effects undermine the quality of data acquisition and association for VSLAM, by capturing the unilateral ground forces at the robot’s feet, and design a system that mitigates their impact.We evaluate our framework on both open and closed-loop bipedal gaits, using a low-cost humanoid platform, and demonstrate that it outperforms kinematic odometry and state-of-the-art dense RGB-D VSLAM methods, by continuously localizing the robot, even in the face of highly irregular and unstable motions.

Contemporary research in human-machine symbiosis has mainly concentrated on enhancing relevant sensory, perceptual, and motor capacities, assuming short-term and nearly momentary interaction sessions. Still, human-machine confluence encompasses an inherent temporal dimension that is typically overlooked. The present work shifts the focus on the temporal and long-lasting aspects of symbiotic human-robot interaction (sHRI). We explore the integration of three time-aware modules, each one focusing on a diverse part of the sHRI timeline. Specifically, the Episodic Memory considers past experiences, the Generative Time Models estimate the progress of ongoing activities, and the Daisy Planner devices plans for the timely accomplishment of goals. The integrated system is employed to coordinate the activities of a multi-agent team. Accordingly, the proposed system (i) predicts human preferences based on past experience, (ii) estimates performance profile and task completion time, by monitoring human activity, and (iii) dynamically adapts multi-agent activity plans to changes in expectation and Human-Robot Interaction (HRI) performance. The system is deployed and extensively assessed in real-world and simulated environments. The obtained results suggest that building upon the unfolding and the temporal properties of team tasks can significantly enhance the fluency of sHRI.

Contact detection is an important topic in contemporary humanoid robotic research. Up to date control and state estimation schemes readily assume that feet contact status is known in advance. In this work, we elaborate on a broader question: in which gait phase is the robot currently in? We introduce an unsupervised learning framework for gait phase estimation based solely on proprioceptive sensing, namely joint encoder, inertial measurement unit and force/torque data. Initially, a meaningful physical explanation on data acquisition is presented. Subsequently, dimensionality reduction is performed to obtain a compact low-dimensional feature representation followed by clustering into three groups, one for each gait phase. The proposed framework is qualitatively and quantitatively assessed in simulation with ground-truth data of uneven/rough terrain walking gaits and insights about the latent gait phase dynamics are drawn. Additionally, its efficacy and robustness is demonstrated when incorporated in leg odometry computation. Since our implementation is based on sensing that is commonly available on humanoids today, we release an open-source ROS/Python package to reinforce further research endeavors.