Robot vision and human-robot interaction, with particular focus on geometric algebra.
- Lasenby, A., Lasenby, J., Matsantonis C. (2022) Reconstructing a Rotor from Initial and Final Frames using Characteristic Multivectors: with applications in Orthogonal Transformations. Advances in Applied Clifford Algebras.
- RDC Course: Geometric Algebra and its applications to robotics and how it can be used for the reconstruction of a 3D scene structure from images; an active research topic in computer vision.
- AgriFoRwArdS CDT Annual Conference (2022): New techniques for problems in vision and robotics using Geometric Algebr.
Other Activities and Outputs
- Took part in the AgriFoRwArdS Summer School 2021 resulting in a co-authored presentation at the AgriFoRwArdS CDT Annual Conference 2021: Automatic Detection of Black Rot in Images of Grapes (in collaboration with Mohammed Terry-Jack, Grey Churchill, YoonJu Cho, Callum Lennox, Haihui Yan).
- University of Cambridge IB A4 Lab Demonstrator.
Previously my study has focused on the areas of mathematics, having obtained a PgC in Mathematics from the Open University and an MSc in Applied Mathematics from Herriot-Watt University. I have already taken part in a series of research activities in the fields of machine learning and sustainability. I would like to investigate the cross-section between mathematics and robotics and devise new computational tools to table challenging problems in robotic manipulation. I am looking forward to applying my knowledge to the areas of robot vision and human-robot interaction, with particular focus on geometric algebra.
Investigation of the use of Geometric Algebra for robotics kinematic and dynamic analysis with applications in selective harvesting
In this project we will investigate the use of the mathematical framework of Geometric Algebra (GA) in 3D reconstruction based on an autonomous strawberry picking system. Conventionally, linear and vector algebra are used to investigate the challenges arising in 3D reconstruction and pose estimation. However, the solutions are often non-geometric and involve symbolic manipulation to produce derivatives. Using GA, we get a common algebraic framework in which vision and kinematics can be expressed. We aim to produce a robust framework for solving the problems arising in stereo vision for 3D reconstruction, and motion estimation of omni-drive fruit picking systems via advanced GA models.
Design and implementation of a machine vision system to promote precision agriculture innovation using novel Geometric Algebra techniques.
The aim of the project is to develop state of the art autonomous robots (articulated or mobile) for agricultural production systems. This is a complex problem given the varying parameters of any given situation: changing workspace, the robot’s kinematic constraints, variation in the input sensor data.
This research is being carried out under the supervision of Prof Joan Lasenby.