Nur Aisyah Syafinaz Suarin

Nur Aisyah Syafinaz Suarin

Auto-generated profile, based on publicly available information

Nur Aisyah Syafinaz Suarin

Perceptive AI Laboratory, MIMOS Berhad, Bukit Jalil, Kuala Lumpur, Malaysia

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
22 September 2022 in Chemometrics and Intelligent Laboratory Systems
Verified
Journal Article
Closed access
XYXien Yin Yap
Xien Yin Yap
KCKim Seng Chia
Kim Seng Chia
NSNur Aisyah Syafinaz Suarin
Nur Aisyah Syafinaz Suarin

16 April 2024 in Knowledge-Based Systems
Verified
Journal Article
Closed access
NSNur Aisyah Syafinaz Suarin
Nur Aisyah Syafinaz Suarin
KCKim Seng Chia
Kim Seng Chia
SFSiti Fatimah Zaharah Mohamad Fuzi
Siti Fatimah Zaharah Mohamad Fuzi

6 March 2025 in IEEE Access
Verified
Journal Article
Open access
AJAmirul Jamaludin
Amirul Jamaludin
SRSyaimaa’ Solehah Mohd Radzi
Syaimaa’ Solehah Mohd Radzi
NYNorhidayah Mohamad Yatim
Norhidayah Mohamad Yatim
NMNazlee Azmeer Masuan
Nazlee Azmeer Masuan
NANorsharimie Mat Adam
Norsharimie Mat Adam
PSPazli Mohd Sulong
Pazli Mohd Sulong
Simultaneous Localization and Mapping (SLAM) algorithms based on visual and light detection and ranging (LiDAR) technologies have shown remarkable capabilities in applications for robot navigation, particularly for mapping and navigating unfamiliar environments. However, the performance of visual and LiDAR based SLAM in challenging settings, such as uneven terrains with complex environmental conditions, remains insufficiently explored. To address this challenge, the feasibility of two visual-based SLAM algorithms, i.e., real-time appearance-based mapping (RTAB-Map) and neuro-inspired 4DoF (degree of freedom) SLAM (NeuroSLAM), and two LiDAR-based SLAM algorithms, i.e., LiDAR inertial odometry via smoothing and mapping (LIO-SAM) and fast LiDAR-inertial odometry (FAST-LIO), were investigated. The algorithms were tested in uneven terrain within a palm oil plantation under varying illuminance conditions (morning and noon) and across trajectories of different lengths, corresponding to map sizes of 400 m (small), 600 m (medium), 1000 m (large), and 1200 m (extra-large). Performance metrics, including distance error and loop closure detection, were used to assess the impact of trajectory length and illuminance. Additionally, the consistency and stability of map estimation were analyzed. Ground-truth data was obtained using the GeoMax Zoom 10 Total Station to ensure high-precision reference measurement. Results revealed that LIO-SAM consistently achieved the lowest distance errors across all map sizes during morning condition of 0.17 m, 1.08 m, 0.52 m, and 0.71 m for small, medium, large, and extra-large maps, respectively. During noon, LIO-SAM also demonstrated superior performance for small, medium, and large map sizes, with distance error values of 0.11 m, 0.25 m, and 0.38 m, respectively. Furthermore, LIO-SAM recorded the highest successful loop closure rates by achieving 75% in the morning and 100% at noon. Overall, the findings highlights the superior performance of LiDAR-based SLAM algorithms over visual-based SLAM approaches in complex, uneven terrains under varying environmental conditions.