Mechatronics Tracking using inertial sensors
Lead Research Organisation:
CRANFIELD UNIVERSITY
Department Name: Cranfield Defence and Security
Abstract
Mechatronics represents the fundamental key area whereby the student can improve the understanding of current systems with free space tracking in a number of applications. Consequently, this underpinning knowledge will result in innovative solutions that overall improve accuracy in a noticeable and practical manner.
People |
ORCID iD |
| John Economou (Primary Supervisor) | |
| Dariusz Maton (Student) |
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/S513623/1 | 01/10/2018 | 30/09/2024 | |||
| 2454420 | Studentship | EP/S513623/1 | 01/10/2020 | 30/09/2024 | Dariusz Maton |
| Description | An algorithm using a form of shallow artificial intelligence (AI) has been designed which uses data from inertial sensors (accelerometers and gyroscopes) to predict the velocity of a user as they walk. The accuracy of the resulting trajectories is competitive with methods using deep learning but is superior from a verification and transparency perspective. This is because the shallow AI used is easily interpretable whereas models learnt using deep learning are not. The work described has been accepted in Sage Journal of Measurement and Control and is awaiting publication. |
| Exploitation Route | The method described in the paper gives insight into the functionality of data-driven methods for velocity estimation. It also highlights the limitations of the method and what researchers need to do to achieve a velocity estimation model that is more general rather than over-fitting to a particular data set. The sensitivity analysis reported in the aforementioned paper demonstrates this problem and paves the way for future work. |
| Sectors | Aerospace, Defence and Marine,Digital/Communication/Information Technologies (including Software),Leisure Activities, including Sports, Recreation and Tourism |