Luca Maria Gambardella

University of Applied Sciences and Arts of Southern Switzerland · Dalle Molle Institute for Artificial Intelligence

This paper addresses the incorporation of problem decomposition skills as an important component of computational thinking (CT) in K-12 computer science (CS) education. Despite the growing integration of CS in schools, there is a lack of consensus on the precise definition of CT in general and decomposition in particular. While decomposition is com...

In this work, we compute the lower bound of the integrality gap of the Metric Steiner Tree Problem (MSTP) on a graph for some small values of number of nodes and terminals. After debating about some limitations of the most used formulation for the Steiner Tree Problem, namely the Bidirected Cut Formulation, we introduce a novel formulation, that we...

Intravital microscopy has revolutionized live-cell imaging by allowing the study of spatial–temporal cell dynamics in living animals. However, the complexity of the data generated by this technology has limited the development of effective computational tools to identify and quantify cell processes. Amongst them, apoptosis is a crucial form of regu...

Intravital microscopy has revolutionized live cell imaging by allowing the study of spatial-temporal cell dynamics in living animals. However, the complexity of the data generated by this technology has limited the development of effective computational tools to identify and quantify cell processes. Amongst them, apoptosis is a crucial form of regu...

Intravital microscopy has revolutionized live cell imaging by allowing the study of spatial-temporal cell dynamics in living animals. However, the complexity of the data generated by this technology has limited the development of effective computational tools to identify and quantify cell processes. Amongst them, apoptosis is a crucial form of regu...

Intravital microscopy has revolutionized live cell imaging by allowing the study of spatial-temporal cell dynamics in living animals. However, the complexity of the data generated by this technology has limited the development of effective computational tools to identify and quantify cell processes. Amongst them, apoptosis is a crucial form of regu...

Collaborative robots are becoming increasingly popular in various industrial contexts, thanks to their ability to safely support workers in heavy and repetitive tasks, without renouncing the human’s decision-making capacity and dexterity. However, their adoption raises several questions about their net effect on cognitive load reduction for workers...

In this work, we propose to classify the empirical hardness of metric Travelling Salesman Problem (TSP) instances by using four features that depend only on the distribution of values in the cost matrix. The first three features are the standard deviation, the skewness, and the Gini index of the cost coefficients. The fourth feature is based on a n...

B-coloring is a theoretical optimization problem on a graph that, on top of being used to model some real-world applications, is exploited by some bounding techniques embedded into solvers for the classical graph coloring problem. This implies that improved solutions for the b-coloring problem have an impact on an even larger pool of practical appl...

This paper discusses a heuristic approach for Team Orienteering Problems with Time Windows. The method we propose takes advantage of a solution model based on a hierarchic generalization of the original problem, which is combined with an Ant Colony System algorithm. Computational results on benchmark instances previously adopted in the literature s...

We propose Hazards&Robots, a dataset for Visual Anomaly Detection in Robotics. The dataset is composed of 324,408 RGB frames, and corresponding feature vectors; it contains 145,470 normal frames and 178,938 anomalous ones categorized in 20 different anomaly classes. The dataset can be used to train and test current and novel visual anomaly detectio...

Live-cell imaging allows the study of apoptosis at cellular level, highlighting morphological hallmarks such as nuclear shrinkage, membrane blebbing, and cell disruption. Identifying the exact location and timing of this process is essential to foster the understanding of its spatial-temporal regulation. However, the analysis of live-cell imaging d...

We consider the problem of building visual anomaly detection systems for mobile robots. Standard anomaly detection models are trained using large datasets composed only of non-anomalous data. However, in robotics applications, it is often the case that (potentially very few) examples of anomalies are available. We tackle the problem of exploiting t...

We consider the task of detecting anomalies for autonomous mobile robots based on vision. We categorize relevant types of visual anomalies and discuss how they can be detected by unsupervised deep learning methods. We propose a novel dataset built specifically for this task, on which we test a state-of-the-art approach; we finally discuss deploymen...

We consider the problem of building visual anomaly detection systems for mobile robots. Standard anomaly detection models are trained using large datasets composed only of non-anomalous data. However, in robotics applications, it is often the case that (potentially very few) examples of anomalies are available. We tackle the problem of exploiting t...

We consider the problem of detecting, in the visual sensing data stream of an autonomous mobile robot, semantic patterns that are unusual (i.e., anomalous) with respect to the robot’s previous experience in similar environments. These anomalies might indicate unforeseen hazards and, in scenarios where failure is costly, can be used to trigger an av...

The ML-Constructive heuristic is a recently presented method and the first hybrid method capable of scaling up to real scale traveling salesman problems. It combines machine learning techniques and classic optimization techniques. In this paper we present improvements to the computational weight of the original deep learning model. In addition, as...

We introduce an approach to train neural network models for visual object localization using a small training set, labeled with ground truth object positions and a large unlabeled one. We assume that the object to be localized emits sound, which is perceived by a microphone rigidly affixed to the camera. This information is used as the target of a...

We consider the task of visually estimating the pose of a human from images acquired by a nearby nano-drone; in this context, we propose a data augmentation approach based on synthetic background substitution to learn a lightweight CNN model from a small real-world training set. Experimental results on data from two different labs proves that the a...

We consider the problem of detecting, in the visual sensing data stream of an autonomous mobile robot, semantic patterns that are unusual (i.e., anomalous) with respect to the robot's previous experience in similar environments. These anomalies might indicate unforeseen hazards and, in scenarios where failure is costly, can be used to trigger an av...

Customers shifting from stationary to online grocery shopping and the decreasing mobility of an ageing population pose major challenges for the stationary grocery retailing sector. To fulfill the increasing demand for online grocery shopping, traditional bricks-and-mortar retailers use existing store networks to offer customers click-and-collect se...

Recent systems applying Machine Learning (ML) to solve the Traveling Salesman Problem (TSP) exhibit issues when they try to scale up to real case scenarios with several hundred vertices. The use of Candidate Lists (CLs) has been brought up to cope with the issues. A CL is defined as a subset of all the edges linked to a given vertex such that it co...

This paper introduces a computational method for generating metric Travelling Salesperson Problem (TSP) instances having a large integrality gap. The method is based on the solution of an NP-hard problem, called IH-OPT, that takes in input a fractional solution of the Subtour Elimination Problem (SEP) on a TSP instance and compute a TSP instance ha...

Customers shifting from stationary to online grocery shopping and the decreasing mobility of an ageing population pose major challenges for the stationary grocery retailing sector. To fulfill the increasing demand for online grocery shopping, traditional bricks-and-mortar retailers use existing store networks to offer customers click-and-collect se...

Recent systems applying Machine Learning (ML) to solve the Traveling Salesman Problem (TSP) exhibit issues when they try to scale up to real case scenarios with several hundred vertices. The use of Candidate Lists (CLs) has been brought up to cope with the issues. The procedure allows to restrict the search space during solution creation, consequen...

The ML-Constructive heuristic is a recently presented method and the first hybrid method capable of scaling up to real scale traveling salesman problems. It combines machine learning techniques and classic optimization techniques. In this paper we present improvements to the computational weight of the original deep learning model. In addition, as...

We propose a general self-supervised learning approach for spatial perception tasks, such as estimating the pose of an object relative to the robot, from onboard sensor readings. The model is learned from training episodes, by relying on: a continuous state estimate, possibly inaccurate and affected by odometry drift; and a detector, that sporadica...

Many emerging applications of nano-sized unmanned aerial vehicles (UAVs), with a few form-factor, revolve around safely interacting with humans in complex scenarios, for example, monitoring their activities or looking after people needing care. Such sophisticated autonomous functionality must be achieved while dealing with severe constraints in pay...

We propose a general self-supervised approach to learn neural models that solve spatial perception tasks, such as estimating the pose of an object relative to the robot, from onboard sensor readings. The model is learned from training episodes, by relying on: a continuous state estimate, possibly inaccurate and affected by odometry drift; and a det...

Artificial intelligence-powered pocket-sized air robots have the potential to revolutionize the Internet-of-Things ecosystem, acting as autonomous, unobtrusive, and ubiquitous smart sensors. With a few cm$^{2}$ form-factor, nano-sized unmanned aerial vehicles (UAVs) are the natural befit for indoor human-drone interaction missions, as the pose esti...

When learning models for real-world robot spatial perception tasks, one might have access only to partial labels: this occurs for example in semi-supervised scenarios (in which labels are not available for a subset of the training instances) or in some types of self-supervised robot learning (where the robot autonomously acquires a labeled training...

The Orienteering Problem is a routing problem aiming at selecting a subset of a given set of customers to be visited within a given time budget, so that a total revenue is maximized. Multiple variants of the problem have been studied. The Probabilistic Orienteering Problem is one of these variants, where customers will require a visit according to...

We propose an industrial measurement and inspection system for steel workpieces eroded by electrical discharge machining, which uses deep neural networks for surface roughness estimation and defect detection. Specifically, a convolutional neural network (CNN) is used as a regressor in order to obtain steel surface roughness and a CNN based on spati...

A comprehensive literature on the Traveling Salesman Problem (TSP) is available, and this problem has become a valuable benchmark to test new heuristic methods for general Combinatorial Optimisation problems. For this reason, recently developed Deep Learning-driven heuristics have been tried on the TSP. These Deep Learning frameworks use the city c...

We present an approach for controlling the position of a quadrotor in 3D space using pointing gestures; the task is difficult because it is in general ambiguous to infer where, along the pointing ray, the robot should go. We propose and validate a pragmatic solution based on a push button acting as a simple additional input device which switches be...

We present a system which allows an operator to land a quadrotor on a precise spot in its proximity by only using pointing gestures; the system has very limited requirements in terms of robot capabilities, relies on an unobtrusive bracelet-like device worn by the operator, and depends on proven, field-ready technologies. During the interaction, the...

We introduce a novel approach to long-range path planning that relies on a learned model to predict the outcome of local motions using possibly partial knowledge, which is trained from a dataset of trajectories acquired in a self-supervised way. Sampling-based path planners use this component to evaluate edges to be added to the planning tree. We i...

In-situ metrology is essential for closed-loop control of machining processes to achieve zero-defect manufacturing: in this context, using inexpensive industrial cameras integrated in machine tools is a widespread solution for dimensional measurements, but has not yet been adopted for measuring surface characteristics such as roughness. This task i...

We present a system which allows an operator to land a quadrotor on a precise spot in its proximity by only using pointing gestures; the system has very limited requirements in terms of robot capabilities, relies on an unobtrusive bracelet-like device worn by the operator, and depends on proven, field-ready technologies. During the interaction, the...

Introduction: The Tourist Trip Design Problem is a variant of a route-planning problem for tourists interested in multiple points of interest. Each point of interest has different availability, and a certain satisfaction score can be achieved when it is visited. Objectives: The objective is to select a subset of points of interests to visit within...

The Probabilistic Orienteering Problem (POP) is a variant of the orienteering problem where customers are available with a certain probability. In a previous work, we approximated its objective function value by using a Monte Carlo Sampling method. A heuristic speed-up criterion is considered in the objective function evaluator. In this work we stu...

Robotic technologies, whether they are remotely operated vehicles, autonomous agents, assistive devices, or novel control interfaces, offer many promising capabilities for deployment in real‐world environments. Postdisaster scenarios are a particularly relevant target for applying such technologies, due to the challenging conditions faced by rescue...

We demonstrate a self-supervised approach which learns to detect long-range obstacles from video: it automatically obtains training labels by associating the camera frames acquired at a given pose to short-range sensor readings acquired at a different pose.

We showcase a model to generate a soundscape from a camera stream in real time. The approach relies on a training video with an associated meaningful audio track; a granular synthesizer generates a novel sound by randomly sampling and mixing audio data from such video, favoring timestamps whose frame is similar to the current camera frame; the sema...

Owing to the significance of combinatorial search strategies both for academia and industry, the introduction of new techniques is a fast growing research field these days. These strategies have really taken different forms ranging from simple to complex strategies in order to solve all forms of combinatorial problems. Nonetheless, despite the kind...

We present a system for interaction between co-located humans and mobile robots, which uses pointing gestures sensed by a wrist-mounted IMU. The operator begins by pointing, for a short time, at a moving robot. The system thus simultaneously determines: that the operator wants to interact; the robot they want to interact with; and the relative pose...

We propose a system to control robots in the users proximity with pointing gestures-a natural device that people use all the time to communicate with each other. Our system has two requirements: first, the robot must be able to reconstruct its own motion, e.g. by means of visual odometry; second, the user must wear a wristband or smartwatch with an...

Cooperative Multi-Robot Observation of Multiple Moving Targets (CMOMMT) denotes a class of problems in which a set of autonomous mobile robots equipped with limited-range sensors keep under observation a (possibly larger) set of mobile targets. In the existing literature, it is common to let the robots cooperatively plan their motion in order to ma...

We introduce a general self-supervised approach to predict the future outputs of a short-range sensor (such as a proximity sensor) given the current outputs of a long-range sensor (such as a camera); we assume that the former is directly related to some piece of information to be perceived (such as the presence of an obstacle in a given position),...

We propose a novel approach to establish the relative pose of a mobile robot with respect to an operator that wants to interact with it; we focus on scenarios in which the robot is in the same environment as the operator, and is visible to them. The approach is based on comparing the trajectory of the robot, which is known in the robot's odometry f...

We consider the task of controlling a quadrotor to hover in front of a freely moving user, using input data from an onboard camera. On this specific task we compare two widespread learning paradigms: a mediated approach, which learns an high-level state from the input and then uses it for deriving control signals; and an end-to-end approach, which...

We introduce a general self-supervised approach to predict the future outputs of a short-range sensor (such as a proximity sensor) given the current outputs of a long-range sensor (such as a camera); we assume that the former is directly related to some piece of information to be perceived (such as the presence of an obstacle in a given position),...

We propose a model of artificial emotions for adaptation and implicit coordination in multi-robot systems. Artificial emotions play two roles, which resemble their function in animals and humans: modulators of individual behavior, and means of communication for social coordination. Emotions are modeled as compressed representations of the internal...

Automatic knowledge grounding is still an open problem in cognitive robotics. Recent research in developmental robotics suggests that a robot's interaction with its environment is a valuable source for collecting such knowledge about the effects of robot's actions. A useful concept for this process is that of an affordance, defined as a relationshi...

We propose a learning-based system for detecting when a user performs a pointing gesture, using data acquired from IMU sensors, by means of a 1D convolutional neural network. We quantitatively evaluate the resulting detection accuracy, and discuss an application to a human-robot interaction task where pointing gestures are used to guide a quadrotor...

We detect and localize obstacles in front of a mobile robot by means of a deep neural network that maps images acquired from a forward-looking camera to the outputs of five proximity sensors. The robot autonomously acquires training data in multiple environments; once trained, the network can detect obstacles and their position also in unseen scena...

Thymio is a small, inexpensive, mass-produced mobile robot with widespread use in primary and secondary education. In order to make it more versatile and effectively use it in later educational stages, including university levels, we have expanded Thymio's capabilities by adding off-the-shelf hardware and open software components. The resulting rob...

We present an interactive guided activity to introduce supervised learning by training a deep neural network (treated as a black box) to recognize "rock paper scissors" hand gestures from unconstrained images. The audience is actively involved in acquiring a varied and representative dataset, on which the rest of the activity is based. Covered conc...

Context: Combinatorial testing strategies have lately received a lot of attention as a result of their diverse applications. In its simple form, a combinatorial strategy can reduce several input parameters (configurations) of a system into a small set based on their interaction (or combination). In practice, the input configurations of software sys...

We demonstrate an intuitive gesture-based interface for manually guiding a drone to land on a precise spot. Using unobtrusive wearable sensors, an operator can quickly and accurately maneuver and land the drone after very little training; a preliminary user study on 5 subjects shows that the system compares favorably with a traditional joystick int...

Mobile ground robots operating on unstructured terrain must predict which areas of the environment they are able to pass in order to plan feasible paths. We address traversability estimation as a heightmap classification problem: we build a convolutional neural network that, given an image representing the heightmap of a terrain patch, predicts whe...