Effects of cow reproductive status, parity and lactation stage on behaviour and heavy breathing indications of a commercial accelerometer during hot weather conditions.

Heat stress presents one of the most urgent challenges to modern dairy farming, having major detrimental impacts on cow welfare, health, and production. Understanding the effect of cow factors (reproductive status, parity, and lactation stage) on the physiological and behavioural response to hot weather conditions is essential for the accurate detection and practical application of heat mitigation strategies. To study this, collars with commercial accelerometer-based sensors were fitted on 48 lactation dairy cows to record behaviour and heavy breathing from late spring to late summer. The temperature-humidity index (THI) was calculated from measurements of 8 barn sensors. We found that, above a THI of 84, cows in advanced pregnancy (>90 days) spent more time breathing heavily and less time eating and in low activity than other cows, while cows in early pregnancy (≤90 days) spent less time breathing heavily, more time eating and in low activity. Cows with 3+ lactations showed less time breathing heavily and in high activity and more time ruminating and in low activity than cows with fewer lactations. Although lactation stage interacted significantly with THI on time spent breathing heavily, ruminating, eating, and in low activity, there was no clear indication at which lactation stage cows were more sensitive to heat. These findings show that cow factors affect the cow's physiological and behavioural response to heat, which could be used to provide group-specific heat abatement strategies, thereby improving heat stress management.

A multi-robot deep Q-learning framework for priority-based sanitization of railway stations.

Sanitizing railway stations is a relevant issue, primarily due to the recent evolution of the Covid-19 pandemic. In this work, we propose a multi-robot approach to sanitize railway stations based on a distributed Deep Q-Learning technique. The proposed framework relies on anonymous data from existing WiFi networks to dynamically estimate crowded areas within the station and to develop a heatmap of prioritized areas to be sanitized. Such heatmap is then provided to a team of cleaning robots - each endowed with a robot-specific convolutional neural network - that learn how to effectively cooperate and sanitize the station's areas according to the associated priorities. The proposed approach is evaluated in a realistic simulation scenario provided by the Italian largest railways station: Roma Termini. In this setting, we consider different case studies to assess how the approach scales with the number of robots and how the trained system performs with a real dataset retrieved from a one-day data recording of the station's WiFi network.

Describing the trend of ammonia, particulate matter and nitrogen oxides: The role of livestock activities in northern Italy during Covid-19 quarantine.

Nitrogen oxides (NOx), sulphur oxides (SOx) and ammonia (NH3) are among the main contributors to the formation of secondary particulate matter (PM2.5), which represent a severe risk to human health. Even if important improvements have been achieved worldwide, traffic, industrial activities, and the energy sector are mostly responsible for NOx and SOx release; instead, the agricultural sector is mainly responsible for NH3 emissions. Due to the emergency of coronavirus disease, in Italy schools and universities have been locked down from late February 2020, followed in March by almost all production and industrial activities as well as road transport, except for the agricultural ones. This study aims to analyze NH3, PM2.5 and NOx emissions in principal livestock provinces in the Lombardy region (Brescia, Cremona, Lodi, and Mantua) to evaluate if and how air emissions have changed during this quarantine period respect to 2016-2019. For each province, meteorological and air quality data were collected from the database of the Regional Agency for the Protection of the Environment, considering both data stations located in the city and the countryside. In the 2020 selected period, PM2.5 reduction was higher compared to the previous years, especially in February and March. Respect to February, PM2.5 released in March in the city stations reduced by 19%-32% in 2016-2019 and by 21%-41% in 2020. Similarly, NOx data of 2020 were lower than in the 2016-2019 period (reduction in March respect to February of 22-42% for 2016-2019 and of 43-62% for 2020); in particular, this can be observed in city stations, because of the current reduction in anthropogenic emissions related to traffic and industrial activities. A different trend with no reductions was observed for NH3 emissions, as agricultural activities have not stopped during the lockdown. Air quality is affected by many variables, for which making conclusions requires a holistic perspective. Therefore, all sectors must play a role to contribute to the reduction of harmful pollutants.

Effect of a Biological Additive on Nitrogen Losses from Pig Slurry during Storage.

Additives applied to animal manure slurries can affect the chemical composition and the biological processes of slurries during storage, with possible improvement of their management and reduction of environmental problems. Some new formulations are marketed claiming a nitrogen (N) removal effect due to denitrification, with the consequence of a reduced N content in the manure after storage. This study evaluated the effects of one of these commercial additives (BACTYcomplex) on slurry characteristics and N losses at a commercial piggery. The additive was applied to four different sectors of the piggery, each with an independent under-floor slurry pit; four other sectors served as controls without treatment. Pits were emptied every 4 wk, and the manure was analyzed for total and ammonia-N and total and volatile solids. Slurry samples from the last month of the on-farm assessment were removed and stored thermostatically in vessels external to the piggery. A subsample of slurry that was treated with the additive at the piggery was treated with an additional dose of additive at the beginning of long-term storage. The additive did not change the composition of the slurry during in-house storage (4 wk duration). During the 155 d of external thermostatic storage, the total solids content of treated slurry was reduced by 18% compared with control slurry, but the N content and composition of treated slurry was unaffected. The additive had a positive effect in accelerating the stabilization of the slurry but did not modify N losses.

A Robotic Cognitive Control Framework for Collaborative Task Execution and Learning.

In social and service robotics, complex collaborative tasks are expected to be executed while interacting with humans in a natural and fluent manner. In this scenario, the robotic system is typically provided with structured tasks to be accomplished, but must also continuously adapt to human activities, commands, and interventions. We propose to tackle these issues by exploiting the concept of cognitive control, introduced in cognitive psychology and neuroscience to describe the executive mechanisms needed to support adaptive responses and complex goal-directed behaviors. Specifically, we rely on a supervisory attentional system to orchestrate the execution of hierarchically organized robotic behaviors. This paradigm seems particularly effective not only for flexible plan execution but also for human-robot interaction, because it directly provides attention mechanisms considered as pivotal for implicit, non-verbal human-human communication. Following this approach, we are currently developing a robotic cognitive control framework enabling collaborative task execution and incremental task learning. In this paper, we provide a uniform overview of the framework illustrating its main features and discussing the potential of the supervisory attentional system paradigm in different scenarios where humans and robots have to collaborate for learning and executing everyday activities.

Development of a New Wearable 3D Sensor Node and Innovative Open Classification System for Dairy Cows' Behavior.

Monitoring dairy cattle behavior can improve the detection of health and welfare issues for early interventions. Often commercial sensors do not provide researchers with sufficient raw and open data; therefore, the aim of this study was to develop an open and customizable system to classify cattle behaviors. A 3D accelerometer device and host-board (i.e., sensor node) were embedded in a case and fixed on a dairy cow collar. It was developed to work in two modes: (1) acquisition mode, where a mobile application supported the raw data collection during observations; and (2) operating mode, where data was processed and sent to a gateway and on the cloud. Accelerations were sampled at 25 Hz and behaviors were classified in 10-min windows. Several algorithms were trained with the 108 h of behavioral data acquired from 32 cows on 3 farms, and after evaluating their computational/memory complexity and accuracy, the Decision Tree algorithm was selected. This model detected standing, lying, eating, and ruminating with an average accuracy of 85.12%. The open nature of this system enables for the addition of other functions (e.g., real-time localization of cows) and the integration with other information sources, e.g., microenvironment and air quality sensors, thereby enhancing data processing potential.

Dairy Cow Behavior Is Affected by Period, Time of Day and Housing.

Dairy cow behavior is affected by external and endogenous factors, including time of year, barn microclimate, time of day and housing. However, little is known about the combined effects of these factors. Data were collected on eight farms in Northern Italy during summer, winter and a temperate season. The temperature-humidity index (THI) was recorded using environmental sensors, whereas cow behavior was monitored using leg accelerometers and cameras. Period, time of day and their interaction all significantly affected lying, standing and feeding behavior. However, although THI had a significant negative effect on lying and a positive effect on standing during daytime (all p < 0.001), during nighttime, it only had a significant negative effect on lying duration and mean lying bout duration (p < 0.001 for both). There was also significant variation between farms in all behavioral parameters, as well as interactions with period and time of day. For instance, farm differences in lying duration were more pronounced during daytime than during nighttime. These findings show how housing can interact with other factors, such as period of the year and time of day, and illustrate the influence of barn structure and farm management on cow behavior and, consequently, their welfare.

A Survey of Dairy Cattle Behavior in Different Barns in Northern Italy.

Due to its increasing pressure on dairy cows, studies that investigate how to cope with heat stress are needed. The heat stress affects multiple aspects of cows' lives, among which their behavior and welfare. In this study, a survey was carried out in eight farms located in Northern Italy to monitor and evaluate the environmental aspects of the barns and the behavioral responses of dairy cows. For one year, three periods were monitored: thermoneutral (T_S), hot (H_S) and cold (C_S) seasons. Temperature and relative humidity were measured by environmental sensors, and lying vs. standing time, number of lying bouts and their average duration were collected by accelerometers. The temperature-humidity index (THI) was quantified inside and outside of the barn. Results show that at the increase of the THI, behavioral adaptations occurred in all the farms, especially with a reduction of lying time and an increase of respiration rate. Four of the eight farms need interventions for improving the cows' welfare. Here, environmental problems should be solved by introducing or improving the efficacy of the forced ventilation or by modifying the barn structure. Monitoring dairy barns with sensors and Precision Livestock Farming techniques can be helpful for future livestock farming to alert farmers on the need for their interventions to respond immediately to unwanted barn living conditions.

Sound Analysis to Predict the Growth of Turkeys.

Protocols for manual weighing of turkeys are not practical on turkey farms because of the large body sizes, heavy weights and flighty nature of turkeys. The sounds turkeys make may be a proxy for bird weights, but the relationship between turkey sounds and bird weights has not been studied. The aim of this study was to correlate peak frequency (PF) of vocalization with the age and weight of the bird and examine the possibility using PF to predict the weight of turkeys. The study consisted of four trials in Egypt. Sounds of birds and their weights were recorded for 11 days during the growth period in each trial. A total 2200 sounds were manually analyzed and labelled by extracting individual and general sounds on the basis of the amplitude and frequency of the sound signal. The PF of vocalizations in each trial, as well as in pooled trails, were evaluated to determine the relationship between PF and the age and weight of the turkey. PF exhibited a highly significant negative correlation with the weight and age of the turkeys showing that PF of vocalizations can be used for predicting the weight of turkeys. Further studies are necessary to refine the procedure.

Review: Environmental impact of livestock farming and Precision Livestock Farming as a mitigation strategy.

This paper reviews the environmental impact of current livestock practices and discusses the advantages offered by Precision Livestock Farming (PLF), as a potential strategy to mitigate environmental risks. PLF is defined as: "the application of process engineering principles and techniques to livestock farming to automatically monitor, model and manage animal production". The primary goal of PLF is to make livestock farming more economically, socially and environmentally sustainable and this can be obtained through the observation, interpretation of behaviours and, if possible, individual control of animals. Furthermore, adopting PLF to support management strategies, may lead to the reduction of the environmental impact of farms. Currently, few studies reported PLF efficacy in reducing the environmental impact, however further studies are necessary to better analyze the actual potential of PLF as a mitigation strategy. Literature shows the potentiality of the application of PLF, as the introduction of PLF in farms can lead to a reduction of Greenhouse gases (GHG) and ammonia (NH3) emission in air, nitrates and antibiotics pollution in water bodies, phosphorus, antibiotics and heavy metals in the soil.

Effect of ammonia stripping and use of additives on separation of solids, phosphorus, copper and zinc from liquid fractions of animal slurries.

To increase the sustainable reuse of animal manure as fertiliser, in many cases suitable treatment techniques are needed to modify the composition and obtain a balanced nutrient content. This study was conducted to evaluate the best strategies to remove solids, P, Cu and Zn, using two additives Ca(OH)2 and Al2(SO4)3, in combination with an ammonia stripping process. The assessment was carried out on five type of liquid fractions derived from the mechanical separation of: raw pig slurry, pig digested slurry, pig digested slurry after ammonia stripping, pig and cattle digested slurry, pig and cattle digested slurry after ammonia stripping. After the addition of the chemicals, the liquid fractions were mixed and then separated using a static filter. The contents of total solids P, Cu and Zn were determined. The additives effectively improved separation efficiencies which depended on the type of slurry and additive used. The P separation efficiencies ranged from 72% to 93% using Al2(SO4)3, and from 20 to 74% using Ca(OH)2. The use of Al2(SO4)3 always had a more consistent effect on the removal efficiencies than Ca(OH)2. The ammonia stripping process, reducing the alkalinity of the digested liquid fractions, facilitated a higher concentration of elements in the separated fraction. With the addition of Al2(SO4)3 to digestate after stripping the concentration of P, Cu and Zn in the solid fraction generally increased when compared to the same liquid fraction without stripping. The addition of Ca(OH)2 might be effective in removing P before the stripping process with the additional benefit to raise pH and improve the ammonia removal efficiency. These findings indicate that solid-liquid separation of animal manure slurries, assisted by chemical additives and coupled with ammonia stripping, can be a viable option for improving the sustainable use of animal manure as a fertiliser.

Effects of Feeding Frequency on the Lying Behavior of Dairy Cows in a Loose Housing with Automatic Feeding and Milking System.

Management systems in modern dairy farms is an important issue in relation to animal comfort and welfare. The objective of this study was to determine the effect of feed delivery frequency on the behavior patterns, visits to an automatic milking system (AMS) and on milk production of lactating dairy cows. The study was conducted on a commercial dairy farm with automatic feeding and milking systems. Feeding treatments consisted of two different frequencies, high feed delivery frequency (11 deliveries per day) and low feed delivery frequency (six deliveries per day). Lying behavior of 20 dairy cows was electronically monitored. The results obtained showed that 11 deliveries per day feed delivery frequency decreases the number of long-duration lying bouts, which may indicate that a very high feeding frequency disturbs the cows during their resting periods and thus influences both animal comfort and milk production. High feeding frequency may disturb the duration of lying bouts and alter the pattern of lying behavior throughout the day, affecting mainly the lying time during the 60 min before and following the provision of fresh feed. Delivering feed at a low frequency allow cows to distribute more evenly their lying time over the course of the day and improve their utilization of an AMS.

Effects of Climatic Conditions on the Lying Behavior of a Group of Primiparous Dairy Cows.

Currently, lying behavior can be assessed using continuous observations from sensors (e.g., accelerometers). The analysis of digital data deriving from accelerometers is an effective tool for studying livestock behaviors. Despite the large interest in the lying behavior of dairy cows, no reference was found in literature regarding the prediction of lying behavior as a function of the interaction of environmental parameters. The present study aimed to evaluate the influence of climatic conditions (temperature-humidity index, solar radiation, air velocity and rainfalls) on the lying behavior of a group of primiparous dairy cows, using data from accelerometers, and develop a prediction model to identify and predict the lying behavior of dairy cows as a function of the effects of environmental conditions. Results from the. GLM Procedure (SAS) showed that the model was highly significant (p < 0.001) and the r2 was 0.84. All of the effects in the model resulted in being highly significant (p < 0.001). This model, if validated properly, could be a valid early warning system to identify any deviation from the expected behavior, and to assess the effectiveness of thermal stress mitigation strategies.

Nitrogen removal from digested slurries using a simplified ammonia stripping technique.

This study assessed a novel technique for removing nitrogen from digested organic waste based on a slow release of ammonia that was promoted by continuous mixing of the digestate and delivering a continuous air stream across the surface of the liquid. Three 10-day experiments were conducted using two 50-L reactors. In the first two, nitrogen removal efficiencies were evaluated from identical digestates maintained at different temperatures (30°C and 40°C). At the start of the first experiment, the digestates were adjusted to pH 9 using sodium hydroxide, while in the second experiment pH was not adjusted. The highest ammonia removal efficiency (87%) was obtained at 40°C with pH adjustment. However at 40°C without pH adjustment, removal efficiencies of 69% for ammonia and 47% for total nitrogen were obtained. In the third experiment two different digestates were tested at 50°C without pH adjustment. Although the initial chemical characteristics of the digestates were different in this experiment, the ammonia removal efficiencies were very similar (approximately 85%). Despite ammonia removal, the pH increased in all experiments, most likely due to carbon dioxide stripping that was promoted by temperature and mixing. The technique proved to be suitable for removing nitrogen following anaerobic digestion of livestock manure because effective removal was obtained at natural pH (≈8) and 40°C, common operating conditions at typical biogas plants that process manure. Furthermore, the electrical energy requirement to operate the process is limited (estimated to be 3.8kWhm-3digestate). Further improvements may increase the efficiency and reduce the processing time of this treatment technique. Even without these advances slow-rate air stripping of ammonia is a viable option for reducing the environmental impact associated with animal manure management.

The role of intrinsic motivations in attention allocation and shifting.

The concepts of attention and intrinsic motivations are of great interest within adaptive robotic systems, and can be exploited in order to guide, activate, and coordinate multiple concurrent behaviors. Attention allocation strategies represent key capabilities of human beings, which are strictly connected with action selection and execution mechanisms, while intrinsic motivations directly affect the allocation of attentional resources. In this paper we propose a model of Reinforcement Learning (RL), where both these capabilities are involved. RL is deployed to learn how to allocate attentional resources in a behavior-based robotic system, while action selection is obtained as a side effect of the resulting motivated attentional behaviors. Moreover, the influence of intrinsic motivations in attention orientation is obtained by introducing rewards associated with curiosity drives. In this way, the learning process is affected not only by goal-specific rewards, but also by intrinsic motivations.