Development of human-machine interface for mixed-swarm robot team

A HMI wearable with control scheme was developed to interface with a MSRD element for high intensity military surveillance operations in urban environments. Research reveals that MSRD technology is being conceptualised for real-world applications, and computer vision frameworks are utilised to detec...

Full description

Saved in:
Bibliographic Details
Main Author: Nur Amirah Binte Rhymie
Other Authors: Li King Ho Holden
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2023
Subjects:
Online Access:https://hdl.handle.net/10356/167647
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-167647
record_format dspace
spelling sg-ntu-dr.10356-1676472023-06-03T16:50:38Z Development of human-machine interface for mixed-swarm robot team Nur Amirah Binte Rhymie Li King Ho Holden School of Mechanical and Aerospace Engineering Temasek Laboratories @ NTU Tan Yan Hao HoldenLi@ntu.edu.sg, yh.tan@ntu.edu.sg Engineering::Mechanical engineering A HMI wearable with control scheme was developed to interface with a MSRD element for high intensity military surveillance operations in urban environments. Research reveals that MSRD technology is being conceptualised for real-world applications, and computer vision frameworks are utilised to detect and track targets. A decision matrix was used to decide on the eventual HMI and MSRD element, by considering factors that influence development and on-the-ground use. A computer vision framework was also implemented to simulate surveillance and threat detection. Eventually, an arm guard with HMI controls and a JETANK were developed to provide the human operator with real-time information and increased awareness via motion detection algorithm and autonomy over decision-making. The system was evaluated through experiments which involved a circuit to navigate the JETANK with a gesture sensing control scheme using APDS9960 module and employing test objects to analyse detectability rates for the motion detection algorithm. State-of-the-art gesture sensing using computer vision framework has a success rate of 91% as researched by Anand et al, and the control scheme achieved standard success rates. Motion detection algorithm results also produced comparative success rates to Anand et al’s study, suggesting that the algorithm is reliable in detecting camouflaged moving objects. Future developments to the system may include experimenting with the optimal latency for the APDS9960 module and integrating additional features to the motion detection algorithm to further improve the system’s performance. Overall, the presence of a human operator in MSRD teams might be able to provide better situational awareness and operational performance with further developments. Bachelor of Engineering (Mechanical Engineering) 2023-05-30T07:58:47Z 2023-05-30T07:58:47Z 2023 Final Year Project (FYP) Nur Amirah Binte Rhymie (2023). Development of human-machine interface for mixed-swarm robot team. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/167647 https://hdl.handle.net/10356/167647 en C170 application/pdf Nanyang Technological University
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Mechanical engineering
spellingShingle Engineering::Mechanical engineering
Nur Amirah Binte Rhymie
Development of human-machine interface for mixed-swarm robot team
description A HMI wearable with control scheme was developed to interface with a MSRD element for high intensity military surveillance operations in urban environments. Research reveals that MSRD technology is being conceptualised for real-world applications, and computer vision frameworks are utilised to detect and track targets. A decision matrix was used to decide on the eventual HMI and MSRD element, by considering factors that influence development and on-the-ground use. A computer vision framework was also implemented to simulate surveillance and threat detection. Eventually, an arm guard with HMI controls and a JETANK were developed to provide the human operator with real-time information and increased awareness via motion detection algorithm and autonomy over decision-making. The system was evaluated through experiments which involved a circuit to navigate the JETANK with a gesture sensing control scheme using APDS9960 module and employing test objects to analyse detectability rates for the motion detection algorithm. State-of-the-art gesture sensing using computer vision framework has a success rate of 91% as researched by Anand et al, and the control scheme achieved standard success rates. Motion detection algorithm results also produced comparative success rates to Anand et al’s study, suggesting that the algorithm is reliable in detecting camouflaged moving objects. Future developments to the system may include experimenting with the optimal latency for the APDS9960 module and integrating additional features to the motion detection algorithm to further improve the system’s performance. Overall, the presence of a human operator in MSRD teams might be able to provide better situational awareness and operational performance with further developments.
author2 Li King Ho Holden
author_facet Li King Ho Holden
Nur Amirah Binte Rhymie
format Final Year Project
author Nur Amirah Binte Rhymie
author_sort Nur Amirah Binte Rhymie
title Development of human-machine interface for mixed-swarm robot team
title_short Development of human-machine interface for mixed-swarm robot team
title_full Development of human-machine interface for mixed-swarm robot team
title_fullStr Development of human-machine interface for mixed-swarm robot team
title_full_unstemmed Development of human-machine interface for mixed-swarm robot team
title_sort development of human-machine interface for mixed-swarm robot team
publisher Nanyang Technological University
publishDate 2023
url https://hdl.handle.net/10356/167647
_version_ 1772829023930417152