Alessandro Manzi

Development of a Socially Believable Multi-Robot Solution from Town to Home

Abstract Technological advances in the robotic and ICT fields represent an effective solution to address specific societal problems to support ageing and independent life. One of the key factors for these technologies is that they have to be socially acceptable and believable to the end-users. This paper aimed to present some technological aspects that have been faced to develop the Robot-Era system, a multi-robotic system that is able to act in a socially believable way in the environments daily inhabited by humans, such as urban areas, buildings and homes. In particular, this paper focuses on two services—shopping delivery and garbage collection—showing preliminary results on experiments conducted with 35 elderly people. The analysis adopts an end-user-oriented perspective, considering some of the main attributes of acceptability: usability, attitude, anxiety, trust and quality of life.

The DustBot system : using mobile robots to monitor pollution in pedestrian area

The EU project DustBot addresses urban hygiene. Two types of robots were designed, the DustClean robot to autonomously clean pedestrian areas, and the DustCart robot for door-to-door garbage collection. Three prototype robots were built and equipped with electronic noses so as to enable them to collect environmental data while performing their urban hygiene tasks. Essentially, the robots act as a mobile, wireless node in a sensor network. In this paper we give an overview of the DustBot platform focusing on the Air Monitoring Module (AMM). We describe the data flow between the robots through the ubiquitous network to a gas distribution modelling server, where a gas distribution model is computed. We describe the Kernel DM+V algorithm, an approach to create statistical gas distribution models in the form of predictive mean and variance discretized onto a grid map. Finally we present and discuss results obtained with the DustBot AMM during experimental trials performed in outdoor public places: a courtyard in Pontedera, Italy and a pedestrian square in Orebro, Sweden.

DustCart, an autonomous robot for door-to-door garbage collection: From DustBot project to the experimentation in the small town of Peccioli

We report on the design and the experimental results of DustCart, a wheeled autonomous robot for door-to-door garbage collection. DustCart is able to navigate in urban environments avoiding static and dynamic obstacles and to interact with human users. The robot is managed by an Ambient Intelligence system (AmI) through a wireless connection: it navigates to collect garbage bags to the houses of users and then moves to discharge the collected waste to a dedicated area. The architecture, navigation and localization systems are described along with the results achieved in different urban sites. In particular, a localization approach based on optical beacons was used and guaranteed position errors sufficient for a safe robot navigation. We report also the first results of a long-term experimentation of the DustCart robot in Peccioli, a small town of Tuscany (Italy). This can be considered as a first step in using robotics in the everyday life of a real town for providing a real service.

Improving Domiciliary Robotic Services by Integrating the ASTRO Robot in an AmI Infrastructure

This work describes the ECHORD Experiment ASTROMOBILE, a project aimed to design, develop and test a system for favourable independent living, improved quality of life and efficiency of care for senior citizens in domestic environments. The system, composed of a mobile robotic platform (called ASTRO) and an Ambient Intelligent Infrastructure that actively cooperated between them and with the end-user, was designed and implemented with a user-centred design approach, involving different stakeholders. The system was designed to deliver services to users, like drug delivery, stand support, reminding, info-entertainment. The design took advantages of the integration of robotic platforms with smart environments, to provide to users higher quality and localization based services. Senior end-users were involved in the experimentation of the system in the DomoCasa Living Lab and feedbacks were gathered for the technology assessment. Particularly, this paper demonstrates the general feasibility of the ASTROMOBILE system and thanks to users feedbacks its acceptability and usability.

The HydroNet ASV, a Small-Sized Autonomous Catamaran for Real-Time Monitoring of Water Quality: From Design to Missions at Sea

In this paper, we describe the design, the development, and the sea trials of a novel small-sized autonomous surface vehicle (ASV) designed for monitoring the coastal water quality. The vehicle is characterized by the capability of measuring hydrocarbon and heavy metal concentrations directly onboard by means of custom-made miniaturized sensors. This capability, novel for an ASV, is combined with a winch-based sampling system specifically designed for small-sized vehicles. The sampling system can collect water samples up to 50 m in depth and measure the physical/water quality parameters of the water column. With these two features, the HydroNet ASV provides an autonomous, practical, real-time monitoring system, conceived to complement the current water monitoring practices in which samples are collected by a dedicated boat and analyzed in specialized laboratories at a later stage. The design process had the aim of realizing a vehicle capable of hosting the sampling system and the custom-made sensors that represent a unique payload in the world of small-sized ASVs. A twofold objective was pursued: realizing an ASV suited for monitoring missions in realistic scenarios (e.g., attention was paid to avoid water sample contamination), at the same time limiting the size for the ease of transportability and deployment. Severe constraints rose from these considerations and were addressed during the realization of the robot such as reduced length/weight (that limit the available space for the sensor payload) and low draft and protected propellers to allow the ASV to navigate in shallow waters with likely floating obstacles such as plastic bags. We report the design process aiming at a tradeoff between ease of transportability (small vehicle), available payload, and navigation performance in terms of achievable speed, endurance, and resistance to environmental disturbances (favored by larger ASV dimensions). We present sea trials of the realized vehicle validating the design choices. In particular, a long-range mission is discussed in which the robot executed a monitoring survey covering autonomously 12.5 km in front of Livorno, Italy, coast.

An autonomous water monitoring and sampling system for small-sized ASV operations

In recent years sensorized autonomous vehicles (either AUVs or ASVs) have been widely utilized for in-situ water measurements. However, collection of water samples in depth by small-sized ASVs, remains difficult due to dimensions and weight constraints. This paper addresses this issue and describes the design and preliminary tests of an autonomous water monitoring and sampling system intended for the use onboard man-portable ASVs. The system is composed of a winch lowering a sampling probe up to 50 meter depth. The sampling probe is able to take measurements along the water column and to collect up to five water samples at different selectable depths. Once the probe is returned onboard, a fluidic system supplies the water to the sensors lodged in the ASV or to containers to transport the samples to laboratories. Particular care has been devoted to avoiding samples contamination by using chemically inert materials. This, also considered the possibility of the need of measuring low concentrations of heavy metals in the samples. The design aims at realizing a system characterized by low dimensions and weight with a limited power consumption, to be easily installed and used on an autonomous man-portable ASV. Preliminary tests are reported with the sampling system installed in a robotic catamaran belonging to the HydroNet class ASV (see Fig. 1) [1].

Enabling Global Robot Navigation Based on a Cloud Robotics Approach

In the future, social robots will permeate our daily life. An autonomous robot that has to move among different buildings needs to manage huge amount of data, as a consequence it is clear that the configuration of the navigation system becomes hard to manage. This paper presents a system, based on a cloud robotics paradigm, conceived to allow autonomous robots to navigate in indoor environment, which are not known a priori. The environment is divided into sub-maps and all the necessary information and the topological representation of the world, are stored into a remote cloud infrastructure. By means of specific environmental tags, composed of a set of ARTags and QR codes, the robot gets the access to the cloud service and it is able to update its navigation configuration in a dynamic and automatic way. Experiments have been conducted in order to choose an appropriate marker dimension and to demonstrate the feasibility of the proposed procedure.

An Autonomous Water Monitoring and Sampling System for Small-Sized ASVs

In recent years, sensorized autonomous vehicles (either AUVs or ASVs) have been increasingly used for in situ water measurements. However, collection of water samplings at depth by small-sized ASVs and their subsequent physical/chemical analysis onboard remains difficult due to size and weight constraints. This paper addresses this issue and describes the design and testing of an autonomous water monitoring and sampling system intended for operations onboard small-sized man-portable ASVs. The system is designed to collect water samples up to 50 m in depth and to measure physical water parameters along the water column. The system is composed of a probe lowered by a winch measuring physical water parameters and able to collect water samples at different selectable depths. Once the probe is returned onboard, a water distribution system transfers the collected water samples to sensors lodged in the ASV for the monitoring of chemical parameters, or into containers to transport the samples to laboratories on the mainland. The system combines small dimensions, self-cleaning capabilities, low weight, and limited power consumption allowing it to be easily installable and used on an autonomous small-sized ASVs. The sampling system was installed and tested in a robotic small-sized catamaran belonging to the HydroNet ASV class.

Two-person activity recognition using skeleton data

Human activity recognition is an important and active field of research having a wide range of applications in numerous fields including ambient-assisted living (AL). Although most of the researches are focused on the single user, the ability to recognise two-person interactions is perhaps more important for its social implications. This study presents a two-person activity recognition system that uses skeleton data extracted from a depth camera. The human actions are encoded using a set of a few basic postures obtained with an unsupervised clustering approach. Multiclass support vector machines are used to build models on the training set, whereas the X-means algorithm is employed to dynamically find the optimal number of clusters for each sample during the classification phase. The system is evaluated on the Institute of Systems and Robotics (ISR) - University of Lincoln (UoL) and Stony Brook University (SBU) datasets, reaching overall accuracies of 0.87 and 0.88, respectively. Although the results show that the performances of the system are comparable with the state of the art, recognition improvements are obtained with the activities related to health-care environments, showing promise for applications in the AL realm.

A Human Activity Recognition System Based on Dynamic Clustering of Skeleton Data

Human activity recognition is an important area in computer vision, with its wide range of applications including ambient assisted living. In this paper, an activity recognition system based on skeleton data extracted from a depth camera is presented. The system makes use of machine learning techniques to classify the actions that are described with a set of a few basic postures. The training phase creates several models related to the number of clustered postures by means of a multiclass Support Vector Machine (SVM), trained with Sequential Minimal Optimization (SMO). The classification phase adopts the X-means algorithm to find the optimal number of clusters dynamically. The contribution of the paper is twofold. The first aim is to perform activity recognition employing features based on a small number of informative postures, extracted independently from each activity instance; secondly, it aims to assess the minimum number of frames needed for an adequate classification. The system is evaluated on two publicly available datasets, the Cornell Activity Dataset (CAD-60) and the Telecommunication Systems Team (TST) Fall detection dataset. The number of clusters needed to model each instance ranges from two to four elements. The proposed approach reaches excellent performances using only about 4 s of input data (~100 frames) and outperforms the state of the art when it uses approximately 500 frames on the CAD-60 dataset. The results are promising for the test in real context.