Robot-human interaction, fleet robotics and machine perception.
Papers and Presentations
- Davy, J., Ghalmazan, A., Hurst, B., and Owen, A. (2021) Current Advances in Hardware for Agri-Food Manipulation- A review. ICRA Task-Informed Grasping Workshop – III, online. Watch here.
- Larby, D., Gudelis, M., Owen, A., Rajendran., V., and Vayakkattil, S. (2021). Computer vision for quality assessment of apples. AgriFoRwArdS CDT Annual Conference 2021, online.
- Member of the AgriFoRwArdS CDT Equality, Diversity and Inclusion Panel
Since my graduation from the University of Leeds I have spent several years working in the automotive industry. Although trained as a chemical engineer, my first role was in the battery team for a large manufacturer and involved developing an understanding of electrical systems and software. Moving to a smaller consultancy a few years later I worked in a software team and then an electrical systems team. Gradually I have made the move to computer science and I am really excited to be a part of the CDT at Lincoln. Developing robotic or autonomous systems for agri-food is challenging but working in this area has real potential to make an impact.
Since being at Lincoln some of my favourite places to visit are the West Common and The Museum of Lincolnshire Life! I have had a previous life in music and played classical guitar as a solo performer and with various ensembles. I enjoy keeping up with my music and have enjoyed strumming a ukulele to reduce stress when deadlines are approaching!
Multi-Agent Simulation for Pickers and Runners at a Strawberry Farm
This project involves developing a simulation using Python and Mesa (a multi-agent simulation platform) to model a team of workers at a strawberry farm. The procedure of allocating picking tasks to all workers on the farm, especially when there are several fields to pick and the packing station is not stationary, is not a trivial one and so the simulation aims to provide a solution for task allocation which gives an improved efficiency for total picking time as compared to current methods used. The simulation will involve ‘bidding’ for tasks in a very similar way to some of those that can be seen to be effective for robot task allocation in a multi-robot team.
Once the workers have filled their trolleys with punnets of strawberries, they must travel to and then queue at the packing station. I aim to implement a solution for reducing the queuing time for workers by tweaking the task allocation process to include a ‘queuing factor’.