White paper - Agricultural Robotics: The Future of Robotic Agriculture
Duckett, Tom; Pearson, Simon; Blackmore, Simon; Grieve, Bruce; Smith, Melvyn
Melvyn Smith Melvyn.Smith@uwe.ac.uk
Research Centre Director Vision Lab/Prof
Agri-Food is the largest manufacturing sector in the UK. It supports a food chain that generates over £108bn p.a., with 3.9m employees in a truly international industry and exports £20bn of UK manufactured goods. However, the global food chain is under pressure from population growth, climate change, political pressures affecting migration, population drift from rural to urban regions and the demographics of an aging global population. These challenges are recognised in the UK Industrial Strategy white paper and backed by significant investment via a wave 2 Industrial Challenge Fund Investment (“Transforming Food Production: from Farm to Fork”). RAS and associated digital technologies are now seen as enablers of this critical food chain transformation. To meet these challenges, here we review the state of the art of the application of RAS in Agri-Food production and explore research and innovation needs to ensure novel advanced robotic and autonomous reach their full potential and deliver necessary impacts.
The opportunities for RAS range from; the development of field robots that can assist workers by carrying weights and conduct agricultural operations such as crop and animal sensing, weeding and drilling; integration of autonomous system technologies into existing farm operational equipment such as tractors; robotic systems to harvest crops and conduct complex dextrous operations; the use of collaborative and “human in the loop” robotic applications to augment worker productivity and advanced robotic applications, including the use of soft robotics, to drive productivity beyond the farm gate into the factory and retail environment.
RAS technology has the potential to transform food production and the UK has the potential to establish global leadership within the domain. However, there are particular barriers to overcome to secure this vision:
1. The UK RAS community with an interest in Agri-Food is small and highly dispersed. There is an urgent need to defragment and then expand the community.
2. The UK RAS community has no specific training paths or Centres for Doctoral Training to provide trained human resource capacity within Agri-Food.
3. While there has been substantial government investment in translational activities at high Technology Readiness Levels (TRLs), there is insufficient ongoing basic research in Agri-Food RAS at low TRLs to underpin onward innovation delivery for industry.
4. There is a concern that RAS for Agri-Food is not realising its full potential, as the projects being commissioned currently are too few and too small-scale. RAS challenges often involve the complex integration of multiple discrete technologies (e.g. navigation, safe operation, multimodal sensing, automated perception, grasping and manipulation, perception). There is a need to further develop these discrete technologies but also to deliver large-scale industrial applications that resolve integration and interoperability issues. The UK community needs to undertake a few well-chosen large-scale and collaborative “moon shot” projects.
5. The successful delivery of RAS projects within Agri-Food requires close collaboration between the RAS community and with academic and industry practitioners. For example, the breeding of crops with novel phenotypes, such as fruits which are easy to see and pick by robots, may simplify and accelerate the application of RAS technologies. Therefore, there is an urgent need to seek new ways to create RAS and Agri-Food domain networks that can work collaboratively to address key challenges. This is especially important for Agri-Food since success in the sector requires highly complex cross-disciplinary activity. Furthermore, within UKRI most of the Research Councils (EPSRC, BBSRC, NERC, STFC, ESRC and MRC) and Innovate UK directly fund work in Agri-Food, but as yet there is no coordinated and integrated Agri-Food research policy per se.
Our vision is a new generation of smart, flexible, robust, compliant, interconnected robotic systems working seamlessly alongside their human co-workers in farms and food factories. Teams of multi-modal, interoperable robotic systems will self-organise and coordinate their activities with the “human in the loop”. Electric farm and factory robots with interchangeable tools, including low-tillage solutions, novel soft robotic grasping technologies and sensors, will support the sustainable intensification of agriculture, drive manufacturing productivity and underpin future food security.
To deliver this vision the research and innovation needs include the development of robust robotic platforms, suited to agricultural environments, and improved capabilities for sensing and perception, planning and coordination, manipulation and grasping, learning and adaptation, interoperability between robots and existing machinery, and human-robot collaboration, including the key issues of safety and user acceptance.
Technology adoption is likely to occur in measured steps. Most farmers and food producers will need technologies that can be introduced gradually, alongside and within their existing production systems. Thus, for the foreseeable future, humans and robots will frequently operate collaboratively to perform tasks, and that collaboration must be safe. There will be a transition period in which humans and robots work together as first simple and then more complex parts of work are conducted by robots; driving productivity and enabling human jobs to move up the value chain.
Duckett, T., Pearson, S., Blackmore, S., Grieve, B., & Smith, M. (2018). White paper - Agricultural Robotics: The Future of Robotic Agriculture
|Publication Date||Jun 21, 2018|
|Peer Reviewed||Peer Reviewed|
|Keywords||agriculture, robotics, machine vision|
You might also like
Optical Imaging Technology In Colonoscopy - Is There A Role For Photometric Stereo
Eye centre localisation with convolutional neural network based regression
Weed classification in grasslands using convolutional neural networks