Uncertainty over access to seasonal migrant labour is placing the otherwise vibrant UK fresh produce and soft fruit production under unprecedented pressure. The immediate impacts of Brexit and COVID-19 have and are restricting availability to the 69,000 seasonal migrant workers who travel to the UK each year to harvest over £1bn of fruit and vegetables. Robotics and automation technology offers a permanent solution that can disconnect the sector from its labour dependency, whilst also creating high skilled jobs and growth for the UK robotics sector. However, critical challenges remain to develop robotic technology that can be scaled in such a way that it will have a significant impact on the industry. This will require thousands of robots to be deployed over the next few years to perform a wide variety of tasks such as picking, plant protection, inspection, etc. Currently, no agricultural robotics companies are able to scale their robot deployment at a speed that will have a significant impact on the industry within the next couple of years. Robots have proven their potential in agriculture by delivering spraying, mechanical and precision weeding, UV treatment and many other services as a commercial service to growers. The number of robots operating at a commercial scale in open fields globally is very low (not currently anticipated to be more than a few thousand). There is a need for a few very specific technological advances and operational changes before we are in a position to scale this globally. This project will develop the fundamental and underpinning technology that will take us from deploying a few robots to selected growers to a position where we deploy hundreds of robots delivering commercial services at a wider scale across the UK by 2024\. Our project synthesises and demonstrates the outputs of multiple Innovate UK funded research and innovation projects. It represents the last step towards full large-scale deployment of agricultural robots in soft fruit production. While previous projects have developed technology that underpin the services that Saga Robotics are currently providing to commercial customers today at a small scale (~30 robots), this project will focus on developing fundamental technology needed to scale agricultural robots in the UK. The identified technological gaps are related to efficient and cost-effective deployment and operation of robots in the open field without the need for constant supervision by humans.

"The UK soft fruit market is now worth well over £1.3 billion at retail sales values (Source: Kantar) per annum. The UK grows over 160,000 tonnes of fruit and employs 32,000 seasonal and, typically, migrant pickers. Approximately 50% of the total production cost is for labour. The soft fruit industry is extremely concerned with the both the availability of picking and husbandry labour and labour cost inflation. The impact of Brexit is already affecting labour supply and the opportunities to pass on labour cost inflation are weak and challenging. The soft fruit sector is a UK success story and there are still opportunities for expansion and to reduce fruit imports. However, it is very clear that to thrive the industry needs to drive every possible means to improve to labour productivity. Robots for soft fruit production clearly offers great opportunity in the sector. Here we will develop the world's first fleet of multi-modal robots that can carry out a wide variety of tasks in the field. Strawberry production is a complex task, and several different tasks need to be performed throughout the season. This is the first project that reflects this, in that we aim to develop robots that are completely autonomous that can carry out several different tasks in the field. We will build on current research by Saga Robotics, Berry Gardens and the University of Lincoln (UoL), complimenting the team for the first time with the inclusion of the University of Oxford and NIAB. Saga have already demonstrated world leading picking performance (in terms of vision system accuracy and picking speed) for their fruit picking robot. The system will then be integrated into the world leading Thorvald robotic platform that has been developed by Saga with UoL. In addition, we will develop a wide range of other tools that will be integrated with the robots. The robots will autonomously pick up the tools from a tool changer, and they will also charge and dock autonomously. This is a much-needed project that will transform robotic strawberry production from the laboratory bench to a commercially relevant system. The world-wide market for these machines and IP is very significant."

Our vision for future soft fruit farming encompasses fleets of electric robotic and autonomous systems powered by renewable energy that pick, transport, pack fruit whilst gathering data to maximise yield, reduce waste and environmental impacts. Additionally, these technologies underpin industry sustainability by reducing sector reliance on low skilled labour, whilst upskilling the existing workforce. This vision can be delivered by 2025\. However, it's critical the underpinning technologies are demonstrated at scale. This secures a significant KE platform to empower transformation across UK and global supply chains. Our project synthesises and demonstrates the outputs of multiple Innovate UK, Saga Robotics, University of Lincoln, Berry Garden Growers, H2020, UKRI-BBSRC, EPSRC and Research England funded research and innovation projects. It will be the largest known global demonstration of robotic and autonomous (RAS) technologies that fuse multiple application technologies (8) across a single farming system. These will drive resource (carbon, pesticide, water, waste) and labour (fruit picking, handling and logistics) productivity whilst underpinning the transition of one of UK's most vibrant agri food sectors (soft fruit) towards a carbon zero future. Robots will be deployed to optimise physical farm processes, in particular to transport and pick fruit, pack fruit , treat crops to reduce critical pests and diseases (UVC to eliminate powdery mildew / insect pests) and optimise spray use. In addition, they will control the virtual farm by collecting data to monitor crop and fruit growth. Data will be analysed using AI and machine learning technologies, pre-developed at Lincoln, to forecast fruit supply and optimise farm productivity. New insights will be gained in the application of robotic systems across large commercial farming systems, in particular fleet control, charging and logistics operations, optimisation of data processing resources (edge / cloud) and the telecommunications infra- structure required to dispatch large volumes of data. Target deliverables: 1\. Elimination of fossil fuel across all farm logistic operations. 2\. 90% reduction in fungicide use (by UVC) and intrinsic carbon cost. 3\. 30% reduction in packhouse labour,40% reduction in farm labour (plus intrinsic carbon costs associated with people movement etc). 4\. 15% increase in farm productivity (yield per m2) and intrinsic carbon gain. 5\. 20% reduction fruit waste, through accurate forecasting.

We estimate that the UK soft fruit sector employs over 35,000 fruit pickers each day. The roles are low skilled and the sector has a high reliance on EU migrants to fill these posts. The impact of Brexit, plus labour inflation through the national living wage legislation poses a serious and direct threat to the whole sector. The need to drive productivity in the sector is urgent. Of the picking costs we estimate that 30% are for the pickers to simply carry picked fruit from within a greenhouse to an on farm logistics hub. We aim to eliminate or dramatically reduce this cost by developing an autonomous robot which can find a picker and transport picked fruit and empty trays around a farm. The robot will be autonomous, have the capacity to map its environment and to find and safely interact with a picker. The key innovation will be the development of autonomous systems which can work safely and over long periods of time in a complex farm environment. The need to drive productivity in the agri-food sector is a global challenge. This project will directly stimulate new markets and supply chains in the production of advanced autonomous systems to support agricultural producers.

Robotics and Autonomous Systems (RAS) technologies are set to transform global industries. Agri-Food is the largest manufacturing sector in the UK, contributing over £38bn GVA to the UK economy and employing 420,000 people. It supports a food chain (primary farming through to retail), which generates a GVA of £108bn, with 3.9m employees in a truly international industry, with £20bn of exports in 2016. The global food chain cannot be taken for granted: it is under pressure from global population growth, climate change, political pressures affecting migration (e.g. Brexit), population drift from rural to urban regions and the demographics of an aging global population in advanced economies. In addition, jobs in the agri-food sector can be physically demanding, conducted in adverse environments and relatively unrewarding. The opportunity for RAS in Agri-Food is compelling - however, large-scale investment in basic underpinning research is required. We propose to create a CDT that focuses on advanced RAS technologies, which will advance the state of the art by creating the largest global cohort of RAS specialists and leaders focused on the Agri-Food sector. This will include 50 PhD scholarships in projects co-designed with industry to give the UK global leadership in RAS across critical and essential sectors of the world economy, expanding the UK's science and engineering base whilst driving industrial productivity and mitigating the environmental and societal impacts of the currently available solutions. In terms of wider impact, the RAS challenges that need to be overcome in the agri-food sector will have further application across multiple sectors involving field robotics and/or robotics in manufacturing. Studying robots for agriculture and food production together allows us to address fundamental challenges in RAS, while delivering whole supply chain efficiencies and synergies across both sides of the farm gate. Core research themes include autonomous mobility in challenging, often GPS-denied and unstructured environments; manipulation and soft robotics for handling delicate and unstructured food products; sensing and image interpretation in challenging agricultural and manufacturing environments; fleet management systems integrating methods for goal allocation, joint motion planning, coordination and control; and 'co-bots' for maintaining safe human-robot collaboration and interaction in farms and factories. All these themes will be applied across a range of applications in agri-food from soil preparation to selective harvesting and on-site grading, through to food processing, manufacturing and supply chain optimisation. The Centre brings together a unique collaboration of leading researchers from the Universities of Lincoln, Cambridge and East Anglia, located at the heart of the UK agri-food business, together with the Manufacturing Technology Centre, supported by leading industrial partners and stakeholders. The wide-scale engagement with industry (£3.0M committed) and end users in the CDT will enable this basic research to be pushed rapidly towards real-world applications in the agri-food industry. An ongoing training programme will take place throughout the CDT, addressing subject-specific and general scientific and technical skills, agriculture and food manufacturing, Responsible Research and Innovation, entrepreneurship, ethics, EDI, and personal and career development. The programme is supported by excellent facilities, including an agri-robotics field centre with a fleet of state-of-the-art agri-robots; a demonstration farm with arable holdings, glasshouses, polytunnels, and livestock; an experimental food factory with robots for food production and intra-logistics; multiple robotics laboratories; advanced robotic manipulators and mobile robots; advanced sensing, imaging and camera technologies; high-performance computing facilities; and excellent links to industrial facilities and test environments.