The objective of this project is to develop a robust controller and robotic manipulation system that can deal with cluttered and uncertain environments using haptic feedback. Reaching into cluttered and unstructured environments for robots is still a largely unsolved problem. Understanding how we can use tactile or other sensory feedback to exploit the environment to achieve tasks that are not achievable with current techniques with higher precision and faster goal times will greatly improve current manipulation techniques. The project will also involve looking into haptic sensor modalities and designing new sensors such as tactile skins for robotic manipulation and their use in robotic reaching in clutter. Some questions that this project aims to answer are:
- What are some better ways of representing the environment such as contacts within a control system?
- When using tactile feedback, what are some other ways we can use the taxel (tactile pixel) information?
- How can we fuse together multiple sensor types to exploit environmental constraints and use them to our advantage?
- How can we categorise the current frameworks and develop a standardised architecture to be used in future research?
- Design of robot controllers
- Design and integration of robotic force/torque/tactile sensors
- A robot controller which can handle occluded environments
- A framework for integrating force/torque or tactile sensors into a controller
Skills and experience
- Python programming
- Advanced control and modeling techniques
- Robot Dynamics and kinematics
Contact the supervisor for more information.