Conference papers

Permanent URI for this communityhttps://rda.sliit.lk/handle/123456789/4889

Browse

Filters

20252

Advanced Search

Filter by

Settings

Author: search.filters.author.Pandithage, D

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemEmbargo
    LifeBeacon: Offline Emergency Communication and Victim Detection System for Disaster Areas
    (IEEE Computer Society, 2025) Senaratna S.M.T.S; Widanage W.T.N; Muhandiramge M.D.A.D.; Bandara H.K.K.T; Pandithage, D; Abeygunawardhana, P; Wijayasekara, S
    Natural disasters often compromise traditional communication infrastructure, significantly delaying emergency response and coordination efforts. This research presents a novel, disaster-resilient communication system designed to address these challenges through the integration of mobile ad-hoc networking and Wi-Fi-based victim localization. The proposed system comprises of a decentralized, infrastructure-less, self-healing mobile ad-hoc network (MANET) utilizing Bluetooth Low Energy (BLE) and Wi-Fi direct with an encrypted SOS messaging mechanism between end devices and a Wi-Fi probe request-based end device detection mechanism that estimates the location of affected individuals based on their mobile phones. If nodes disconnect, a self-healing algorithm guarantees automatic reconnection and uninterrupted message flow, while the ad-hoc network facilitates low-power, peer-to-peer communication without the need for traditional infrastructure. Both targeted and broadcast notifications are supported by the SOS messaging system, which is encrypted to protect the integrity and security of data. The victim localization system, meantime, makes precise estimates of the population and location of people in the disaster region using trilateration and Received Signal Strength Indicator (RSSI). Experimental evaluations conducted under simulated disaster conditions demonstrate the system's scalability, energy efficiency, and effectiveness in maintaining real-time communication and accurate victim tracking.
  • Thumbnail Image
    ItemEmbargo
    "Cropmaster" - Real-Time Coordination of Multirobot Systems for Autonomous Crop Harvesting: Design and Implementation
    (Institute of Electrical and Electronics Engineers Inc., 2025) Pramod, I; Arachchi, A.M; Rashen, C; Chinthaka, G; Pandithage, D; Gamage, N
    The CropMaster is an autonomous rover system designed to enhance Scotch Bonnet production by improving disease management, crop sorting, autonomous navigation, and real-time environmental monitoring. Equipped with sensors to measure sunlight, humidity, pH, NPK content, and soil moisture, the rover securely transmits analyzed data to a web-based dashboard. LIDAR technology enables efficient autonomous navigation, allowing the rover to move around fields and avoid obstacles. The MQTT protocol facilitates communication between multiple rovers, preventing duplicate measurements and ensuring data is sent to the dashboard for comprehensive data collection across large areas. TensorFlow's machine learning models allow the rover to accurately assess crop health and detect early-stage diseases, followed by automated pesticide and fertilizer application through a spraying system. To maintain reliability, the rover's operations, including data transfer and task execution, are continuously monitored for Quality of Service (QoS). All collected data is stored in the cloud for long-term access. Built with a lightweight aluminum and plastic chassis and robotic arms, the rover is designed for adaptability and operational efficiency, aiming to improve crop management and increase yields across extensive agricultural fields.