-
Enhanced 6D Pose Estimation for Robotic Fruit Picking
Authors:
Marco Costanzo,
Marco De Simone,
Sara Federico,
Ciro Natale,
Salvatore Pirozzi
Abstract:
This paper proposes a novel method to refine the 6D pose estimation inferred by an instance-level deep neural network which processes a single RGB image and that has been trained on synthetic images only. The proposed optimization algorithm usefully exploits the depth measurement of a standard RGB-D camera to estimate the dimensions of the considered object, even though the network is trained on a…
▽ More
This paper proposes a novel method to refine the 6D pose estimation inferred by an instance-level deep neural network which processes a single RGB image and that has been trained on synthetic images only. The proposed optimization algorithm usefully exploits the depth measurement of a standard RGB-D camera to estimate the dimensions of the considered object, even though the network is trained on a single CAD model of the same object with given dimensions. The improved accuracy in the pose estimation allows a robot to grasp apples of various types and significantly different dimensions successfully; this was not possible using the standard pose estimation algorithm, except for the fruits with dimensions very close to those of the CAD drawing used in the training process. Grasping fresh fruits without damaging each item also demands a suitable grasp force control. A parallel gripper equipped with special force/tactile sensors is thus adopted to achieve safe grasps with the minimum force necessary to lift the fruits without any slippage and any deformation at the same time, with no knowledge of their weight.
△ Less
Submitted 25 May, 2023;
originally announced May 2023.
-
Understanding the Relationship between Social Distancing Policies, Traffic Volume, Air Quality, and the Prevalence of COVID-19 Outcomes in Urban Neighborhoods
Authors:
Daniel L. Mendoza,
Tabitha M. Benney,
Rajive Ganguli,
Rambabu Pothina,
Benjamin Krick,
Cheryl S. Pirozzi,
Erik T. Crosman,
Yue Zhang
Abstract:
In response to the COVID-19 pandemic, governments have implemented policies to curb the spread of the novel virus. Little is known about how these policies impact various groups in society. This paper explores the relationship between social distancing policies, traffic volumes and air quality and how they impact various socioeconomic groups. This study aims to understand how disparate communities…
▽ More
In response to the COVID-19 pandemic, governments have implemented policies to curb the spread of the novel virus. Little is known about how these policies impact various groups in society. This paper explores the relationship between social distancing policies, traffic volumes and air quality and how they impact various socioeconomic groups. This study aims to understand how disparate communities respond to Stay-at-Home Orders and other social distancing policies to understand how human behavior in response to policy may play a part in the prevalence of COVID-19 positive cases. We collected data on traffic density, air quality, socio-economic status, and positive cases rates of COVID-19 for each zip code of Salt Lake County, Utah (USA) between February 17 and June 12, 2020. We studied the impact of social distancing policies across three periods of policy implementation. We found that wealthier and whiter zip codes experienced a greater reduction in traffic and air pollution during the Stay-at-Home period. However, air quality did not necessarily follow traffic volumes in every case due to the complexity of interactions between emissions and meteorology. We also found a strong relationship between lower socioeconomic status and positive COVID-19 rates. This study provides initial evidence for social distancing's effectiveness in limiting the spread of COVID-19, while providing insight into how socioeconomic status has compounded vulnerability during this crisis. Behavior restrictions disproportionately benefit whiter and wealthier communities both through protection from spread of COVID-19 and reduction in air pollution. Such findings may be further compounded by the impacts of air pollution, which likely exacerbate COVID-19 transmission and mortality rates. Policy makers need to consider adapting social distancing policies to maximize equity in health protection.
△ Less
Submitted 15 July, 2020;
originally announced August 2020.
-
Absentee and Economic Impact of Low-Level Fine Particulate Matter and Ozone Exposure in K-12 Students
Authors:
Daniel L. Mendoza,
Cheryl S. Pirozzi,
Erik T. Crosman,
Theodore G. Liou,
Yue Zhang,
Jessica J. Cleeves,
Stephen C. Bannister,
William R. L. Anderegg,
Robert Paine III
Abstract:
High air pollution levels are associated with school absences. However, low level pollution impact on individual school absences are under-studied. We modelled PM2.5 and ozone concentrations at 36 schools from July 2015 to June 2018 using data from a dense, research grade regulatory sensor network. We determined exposures and daily absences at each school. We used generalized estimating equations…
▽ More
High air pollution levels are associated with school absences. However, low level pollution impact on individual school absences are under-studied. We modelled PM2.5 and ozone concentrations at 36 schools from July 2015 to June 2018 using data from a dense, research grade regulatory sensor network. We determined exposures and daily absences at each school. We used generalized estimating equations model to retrospectively estimate rate ratios for association between outdoor pollutant concentrations and school absences. We estimated lost school revenue, productivity, and family economic burden. PM2.5 and ozone concentrations and absence rates vary across the School District. Pollution exposure were associated with as high a rate ratio of 1.02 absences per ug/m$^3$ and 1.01 per ppb increase for PM2.5 and ozone, respectively. Significantly, even PM2.5 and ozone exposure below regulatory standards (<12.1 ug/m$^3$ and <55 ppb) was associated with positive rate ratios of absences: 1.04 per ug/m$^3$ and 1.01 per ppb increase, respectively. Granular local measurements enabled demonstration of air pollution impacts that varied between schools undetectable with averaged pollution levels. Reducing pollution by 50% would save $452,000 per year districtwide. Pollution reduction benefits would be greatest in schools located in socioeconomically disadvantaged areas. Exposures to air pollution, even at low levels, are associated with increased school absences. Heterogeneity in exposure, disproportionately affecting socioeconomically disadvantaged schools, points to the need for fine resolution exposure estimation. The economic cost of absences associated with air pollution is substantial even excluding indirect costs such as hospital visits and medication. These findings may help inform decisions about recess during severe pollution events and regulatory considerations for localized pollution sources.
△ Less
Submitted 15 July, 2020;
originally announced July 2020.
-
A Transfer Learning Approach to Cross-Modal Object Recognition: From Visual Observation to Robotic Haptic Exploration
Authors:
Pietro Falco,
Shuang Lu,
Ciro Natale,
Salvatore Pirozzi,
Dongheui Lee
Abstract:
In this work, we introduce the problem of cross-modal visuo-tactile object recognition with robotic active exploration. With this term, we mean that the robot observes a set of objects with visual perception and, later on, it is able to recognize such objects only with tactile exploration, without having touched any object before. Using a machine learning terminology, in our application we have a…
▽ More
In this work, we introduce the problem of cross-modal visuo-tactile object recognition with robotic active exploration. With this term, we mean that the robot observes a set of objects with visual perception and, later on, it is able to recognize such objects only with tactile exploration, without having touched any object before. Using a machine learning terminology, in our application we have a visual training set and a tactile test set, or vice versa. To tackle this problem, we propose an approach constituted by four steps: finding a visuo-tactile common representation, defining a suitable set of features, transferring the features across the domains, and classifying the objects. We show the results of our approach using a set of 15 objects, collecting 40 visual examples and five tactile examples for each object. The proposed approach achieves an accuracy of 94.7%, which is comparable with the accuracy of the monomodal case, i.e., when using visual data both as training set and test set. Moreover, it performs well compared to the human ability, which we have roughly estimated carrying out an experiment with ten participants.
△ Less
Submitted 18 January, 2020;
originally announced January 2020.
-
Manipulation Planning and Control for Shelf Replenishment
Authors:
Marco Costanzo,
Simon Stelter,
Ciro Natale,
Salvatore Pirozzi,
Georg Bartels,
Alexis Maldonado,
Michael Beetz
Abstract:
Manipulation planning and control are relevant building blocks of a robotic system and their tight integration is a key factor to improve robot autonomy and allows robots to perform manipulation tasks of increasing complexity, such as those needed in the in-store logistics domain. Supermarkets contain a large variety of objects to be placed on the shelf layers with specific constraints, doing this…
▽ More
Manipulation planning and control are relevant building blocks of a robotic system and their tight integration is a key factor to improve robot autonomy and allows robots to perform manipulation tasks of increasing complexity, such as those needed in the in-store logistics domain. Supermarkets contain a large variety of objects to be placed on the shelf layers with specific constraints, doing this with a robot is a challenge and requires a high dexterity. However, an integration of reactive grasping control and motion planning can allow robots to perform such tasks even with grippers with limited dexterity. The main contribution of the paper is a novel method for planning manipulation tasks to be executed using a reactive control layer that provides more control modalities, i.e., slipping avoidance and controlled sliding. Experiments with a new force/tactile sensor equipping the gripper of a mobile manipulator show that the approach allows the robot to successfully perform manipulation tasks unfeasible with a standard fixed grasp.
△ Less
Submitted 12 March, 2020; v1 submitted 23 December, 2019;
originally announced December 2019.
