Browsing by Author "Muhammad Muaz Abni Hajar"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
- Some of the metrics are blocked by yourconsent settings
Publication Autonomous UAV-Based Cattle Detection And Counting Using Yolov3 And Deep Sort(Universiti Sains Islam Malaysia, 2021-09-09) ;Muhammad Muaz Abni Hajar ;Noor Hanis Izzuddin Mat Lazim ;Ahmad Rasdan RosdiLiyana RamliUnmanned Aerial Vehicles (UAV) have been one of the most crucial technological advances in this era. One of the sectors that have been taken over gradually by UAV is agriculture. UAV has assisted the farmer in managing farmland autonomously especially in monitoring livestock. This study proposes the cattle detection and counting system based on deep learning via UAV. This detection and counting system is done autonomously for monitoring livestock, specifically cattle. We take advantage also of the assistant of UAVs to replace the “human eye” or real-time image in monitoring the livestock. The deep learning method used in this research is the YOLOv3 and Deep SORT that have been integrated into drone real time images to capture the real-time cattle monitoring system. Keywords: UAV, YOLOv3, Deep SORT, Real-time - Some of the metrics are blocked by yourconsent settings
Publication Leader-Follower Formation Control Of Quadrotors With Collision Avoidance Using Artificial Potential Field(Universiti Sains Islam Malaysia, 2021-09-09) ;Ahmad Rasdan Rosdi ;Noor Hanis Izzuddin Mat Lazim ;Muhammad Muaz Abni HajarLiyana RamliThis paper presents a collision-free virtual leader-follower formation control for a group of quadrotors using artificial potential field (APF) and proportional--derivative (PD) controller. The proposed control approach consists of PD controllers that stabilise the attitude, as well as for the position control. To avoid collision with obstacles, APF algorithm is used at the outer-most loop whereby the collision-free path is fed into the position control. Several simulation experiments were carried out to validate the proposed approach. Results show that the integration of APF to the controller were able to avoid obstacles. Keywords: Formation control, multi-agent system, leader-follower, artificial potential field, unmanned aerial vehicle