This research is focused on robotic grasping and agricultural robotics.
In this project you will use novel multi-perspective camera arrays and learning-based methods, such as reinforcement learning, to enable a robot to choose the direction which optimises the visual information of a scene.
This project aims to develop novel method for robots to move their perception system more intelligently for finding targets in highly occluded and unstructured environments.
There has been a lot of recent interest in machine learning and reinforcement learning which has the potential to enable robots to learn to do complex tasks that are hard to program directly.
Specifically, using novel multi-perspective camera arrays and learning based methods such as reinforcement learning this project aims to enable a robot to choose the direction which optimises the visual information of the scene. This has applications in standard robotic pick and place environments but in more challenging unstructured and natural environments such as agriculture.
Skills and experience
You should skills and experience in:
- electrical PCB design
- Raspberry Pi
- computer vision
- visual serving and control
- robotic inverse kinematics and planning.
Contact the supervisor for more information.