Our food supply is under unprecedented pressure. The FAO (2021) estimates that 20-40% of our crops are lost to pests, at a global industry cost of $70bn each year (WeForum 2021). Therefore integrated pest management (IPM) strategies are increasingly adopted to control outbreaks and minimise economic impact and environmental damage. IPM approaches include use of biological control agents/mating disruption that rely upon precision monitoring to rapidly identify pest outbreaks in order to take timely action, reducing pesticide use by up to 30% and increasing yield by 40% . However, monitoring largely relies on traps with manual checks and counts and therefore monitoring doesn't always occur with the frequency needed to optimise control methods. This project will develop a novel system to help farmers manage pests through precision monitoring without additional labour commitment. Specifically, the project will develop a detection system that distinguishes and counts insect species, implemented as part of a trap or monitoring device adaptable to a range of crops/pests. Traps and monitoring devices will contain innovative chemical lure formulations that improve pest catch. The resultant offering will thereby enhance agricultural productivity, reducing waste and pesticide usage, which reduces the overall carbon footprint of the industry. The innovation supports a transformative change in fruit/vegetable production by enabling farmers to get greater benefit from sustainable pest management approaches.

Our proposed System is a completely non-intrusive face-based (artificial) intelligent monitoring system that can automatically capture welfare and health data for monitoring and management of pigs. Why? Being able to continuously monitor and assess farm animal health and welfare depends on the deployment of practical, valid, and reliable measurement tools. Current on-farm welfare assessment protocols involve daily spot-checks by staff, and periodic spot-checks by veterinarians or quality assurance inspectors. Assessed welfare parameters are often resource-based concerning provision for basic needs, or animal-based, looking mainly at easily-measurable factors such as physical condition. Most are performed at a group level as individual identification can be difficult. Rarely is animal behaviour recorded and even rarer still are measures that can tell us something about the emotional state of the individual animal. How will the system work? Our innovative approach is to measure an animal's emotional state as well as its body condition using a completely non-intrusive face-based, (artificial) intelligent monitoring system. We have already successfully developed machine learning algorithms that identify individuals using facial biometrics and are able to detect changes in facial expression that indicate whether a pig is stressed or not. We have also developed body condition scores and weight estimation using these techniques. Here we propose to combine all of these capabilities into one face based, non-intrusive animal health and welfare monitoring station for use on commercial farms. By employing these state-of-the-art machine learning techniques, our system will offer the capacity for on-going learning about individual animals, and consequently allow for early detection of altered health/welfare, personalised thresholds for intervention, and tailored treatment approaches. Such individualised data recording can be integrated with other measurable parameters, such as individual food and water intake, treatment history, growth and weight gain, which will allow better optimisation of farm production efficiency.

Farmers need accurate knowledge of beef cattle weight in order to manage growth and predict optimal finishing time - ie the point at which growth rate has peaked and the animals generate a larger carbon footprint because FCR decreases. Currently, cattle are weighed at 6-8 week intervals, this is time-consuming and animals do not grow well for the following 2 days after a physical weighing, Ongoing automated weight estimation gives the farmer an easier alternative to weighing. With an automated system that gives accurate daily weights while avoiding the time and disruption of weighing, the farmer can better manage feed, reduce labour costs and reduce emissions. Automated daily weight monitoring enables farmers to precisely manage feed, reducing waste and maximising growth. Selecting appropriate slaughter date reduces age of animals at slaughter, therefore their overall environmental impact, given that ~66% of feed consumed per day is purely for maintenance, rather than growth. There are currently few cost-effective systems which can automatically accurately and regularly measure daily weights, and none known to accurately combine this with a body condition score. Now, the HerdView project will assess the feasibility of adapting Agsenze technology to track daily cow growth and predict weight/grade/fat information for beef cattle. Thereby enabling farmers to better manage feeding while reducing effort and animal distress.

This project will develop and pilot a cloud-based platform that gives inspection bodies access to automated body condition scoring and video information on individual animals, enabling a more thorough welfare inspection without the need for a farm visit. Because of lockdown measures, on-farm inspections can not be implemented and certification bodies such as Red Tractor have enacted some pilot remote inspections, which do not give the same level of animal welfare information as a physical inspection. In addition lockdown measures have caused a backlog of both inspections and training for new inspectors. This technology will augment existing virtual inspection protocols, help to reduce travel time for inspectors, thus helping to reduce the inspection backlog. Making the automated monitoring data available as part of online courses, enhances training modules and enables a greater part of the training to take place online, thus reducing the training backlog. Insurance companies report that the pandemic has created challenges in data gathering for risk management due to a greater degree of uncertainty and a lack of up-to-date information. The envisaged system gives insurance companies access to detailed farm data enabling a greater depth of information on the dairy industry that helps them manage risk during this period of unprecedented uncertainty.

The MothNet project will assess the early stage technical feasibility of a fully automated codling moth monitoring system. The innovative project will gather and analyse data on codling moth features based on a wide range of parameters in order to identify previously unexploited features that can be used to distinguish the species. In addition new long-life pheromone lure designs will be trialled. In this way, the project will enable partners to assess whether the system concept is feasible and should be developed further. This project was conceived in response to feedback from within the fruit growers industry that precision trapping and monitoring systems are urgently needed. Industry requirements are for highly accurate real-time data collection without the need for skilled labour for moth identification and replacement of trap parts - such a system is not currently on the market. Alongside technical feasibility, dedicated work packages explore different business models for the product and routes to adoption in a range of geographical contexts. End-users are engaged from the outset to define target price points and technical requirements. Successful completion of the MothNet project will lead to the development of a precision pest monitoring system that helps to reduce emissions and improve resource use efficiency by enabling rapid targeted response to pest outbreaks that in turn reduces pesticide usage and increases yields from existing agricultural land. Specifically our potential solution to the challenge of pest-predation in the fruit industry is data-driven, using widespread occurrence counts combined with environmental data to define action thresholds. By developing a fully automated system designed around reducing labour for fruit growers, our approach will increase adoption of precision approaches to bridge the productivity gap. Once the initial feasibility of species distinction has been proven, our ambition is to further refine the system to characterise sex and age of codling moth, enabling yet greater precision of control through models based on lifecycle stage of the insects. Following this, adaptation of the system to distinguish and record a range of insect species (both beneficial and harmful) creates an entirely novel device bring about a step-change in agricultural and environmental monitoring.

Honeybees are vital pollinators, both to the agricultural industry and the wider environment. African honeybees are essential for cash crop, such as cashew, production to provide income to smallholder farmers in Africa. The BEE SMART project will test an innovative new commercial model to improve yields and environmental management for smallholder male and female farmers in Ghana. Specifically, the project will pilot low-cost managed pollination services for smallholder cashew farmers to improve yields, thus increasing profits as well as raising awareness of the value of ecosystem services and promoting environmental protection. The proposed strategy is novel for many African countries where pollination services have not yet become a commercial offering, as they have in many other parts of the world, and pollination management has not been integrated into the value chain for smallholders. As such, successful implementation of the project creates a new market in Ghana, benefits smallholder farmers and strengthens collaboration and capacity-building between UK and Ghana.

Chicken and eggs are a major source of both nutrition and income in rural parts of Eastern and Southern Africa. Local bird breeds are often reared in family homes by women who a) directly consume produce or b) sell produce to urban markets. To date, most poultry farming is performed at small scales (<50 birds) but there is a great opportunity to sell more produce to wealthier middle-class people and help women and their families to earn more money. This 18-month project (SmartCoop) aims to support the scale-up of rural poultry production by combining two innovative technologies: self-building chicken coops (sold throughout Southern Africa by Inkukakaya) and bird sound hardware and analytics (developed by UK SME AgsenZe). Together, this project will yield a new class of chicken coop equipped with decision support tools to enable female growers to reduce mortality rates, increase production and upskill through delivery of both agricultural and basic business/commercially-orientated courses (via partnership with NOSA). We expect outputs of this project to impact individuals (increased protein availability and direct revenue), communities (through formation of a female-led co-operative) and nationally.

To realise the functionality of a system for improved dairy herd management using novel animal monitoring technology.

Worldwide demand for meat and animal products is set to increase by c.40% over the next decade; China's total meat production quadrupled in the last 20 years due to rising demand from a rapidly growing population. However, environmental/public health issues are becoming more prominent in China (as with all emerging economies), and sustainable intensification of livestock agriculture is a key concern of Chinese policy-makers & stakeholders. This project merges a team of interdisciplinary experts in animal behaviour, Internet-of-Things (IoT),wearable computing & veterinary diagnostics to develop a highly innovative Smart Wearable IoT platform and Decision Support System for precision livestock farming (with an initial focus on poultry). A fully-networked Smart farmers' boot is proposed to assess animal welfare and farm environment at flock eye-level, allowing ubiquitous, non-obstructive, automated data collection. Guangxi province farm data will standardise animal health and welfare indices for China.This will improve farm productivity, animal welfare, smallholder livelihood and consumer nutrition, contributing to economic development and welfare of the Chinese population.

AflaScope is a cross-disciplinary collaboration to examine the feasibility of using an acoutic separation platform for purification of aflatoxins from crops. Aflatoxins (toxins from storage mould) are a significant threat to food security, particularly in developing nations. Testing & monitoring are vital but, due to complex sample prep, high cost, inaccessibility & lack of information, aflatoxin testing is not thoroughly implement & billions of people are at risk. This innovative project will develop a novel, rapid & chemical- free procedure for extracting & concentrating aflatoxins. When integrated with down-stream diagnostic advances, the extraction platform could enable a low-cost, sensitive, portable test system for on-site aflatoxin monitoring, increasing ease & frequency of testing, & potentially improve decontamination. If successful, the resultant increase in crop value & safety will bring about a step-change in on-farm management.