Integrating advanced reporter nanomaterials and molecules to indicate the 'health' of biomedical materials during manufacture
Lead Research Organisation:
University of Cambridge
Department Name: Engineering
Abstract
This project falls under the category of Robotics of the EPSRC research areas.
With state of the art neural networks performing better at numerous image recognition and classification tasks than human experts, the field of computer vision is transforming to performing increasingly more higher level reasoning. Inspired by advances in human robot interactions where robot arms are lead directly by a human this project will investigate teaching drones certain maneuvers through human speech.
Directing a drone using spoken language is difficult because human speech is intrinsically very dependent on the current situation and relies on numerous unspoken assumptions. While most current research has been focusing on converting natural language directions into navigation this research will use natural language for learning explicit actions. Particularly important for this project will be the problem of relative localization and navigation. Humans think in relative reference frames and the main challenge of this project will be converting relative instructions into a sequence of precise 6DoF drone positions.
With state of the art neural networks performing better at numerous image recognition and classification tasks than human experts, the field of computer vision is transforming to performing increasingly more higher level reasoning. Inspired by advances in human robot interactions where robot arms are lead directly by a human this project will investigate teaching drones certain maneuvers through human speech.
Directing a drone using spoken language is difficult because human speech is intrinsically very dependent on the current situation and relies on numerous unspoken assumptions. While most current research has been focusing on converting natural language directions into navigation this research will use natural language for learning explicit actions. Particularly important for this project will be the problem of relative localization and navigation. Humans think in relative reference frames and the main challenge of this project will be converting relative instructions into a sequence of precise 6DoF drone positions.
Organisations
People |
ORCID iD |
Florian Langer (Student) |
Studentship Projects
Project Reference | Relationship | Related To | Start | End | Student Name |
---|---|---|---|---|---|
EP/R513180/1 | 30/09/2018 | 29/09/2023 | |||
2435439 | Studentship | EP/R513180/1 | 30/09/2020 | 31/03/2024 | Florian Langer |
EP/T517847/1 | 30/09/2020 | 29/09/2025 | |||
2435439 | Studentship | EP/T517847/1 | 30/09/2020 | 31/03/2024 | Florian Langer |