Integrating Wireless Mesh Networking and Object-Oriented Semantic Mapping into UAVs
Disciplines
Electrical and Electronics
Abstract (300 words maximum)
Unmanned Aerial Vehicles (UAVs) are becoming increasingly utilized in applications such as disaster response, surveillance, and extending connectivity. However, reliance on human operators and centralized communication systems introduces risks of human errors, and miscommunication, which has led to serious accidents such as the aircraft collision in DC. Ensuring reliable communication and collision avoidance among UAVs is a major factor in operational success. This research explores how Wireless Mesh Networking (WMN), and Object-Oriented Semantic Mapping could be implemented with UAVs to improve wireless communication and collision avoidance. WMNs are a type of network topology which allows for scalable, decentralized communication with adaptive routing. Object-Oriented Semantic Mapping creates 3D maps providing object-specific data, improving environmental awareness. WMNs enhance UAVs by providing decentralized communication paired with robust connectivity and adaptive routing, allowing UAVs to effectively communicate and maintain connectivity in dynamic environments. Choosing routing and distributed MAC protocols such as AODV and CSMA/CA respectively or others, creates a robust communication network and improves efficiency. Object-Oriented Semantic Mapping enhances UAV navigation by providing object detection and probabilistic mapping, allowing for UAVs to interpret and respond to obstacles effectively. By integrating YOLOv8-based real-time object detection and probabilistic SLAM techniques, UAVs can create detailed maps that can identify obstacles and improve navigation decisions. The integration of these technologies allows for the creation of a UAV network capable of real-time collaboration, and obstacle avoidance. The adjustments prove Wireless Mesh Networking together with Object-Oriented Semantic Mapping enables UAVs to operate with better collision avoidance as well as robust communication. The combination of real-time object detection and probabilistic mapping with optimized data transmission enables UAVs to work better in dynamic conditions while requiring minimal human supervision. Future research aims to perfect these technologies because their applications in self-driving cars and aircraft designs contributes to more reliable autonomous systems.
Academic department under which the project should be listed
SPCEET - Electrical and Computer Engineering
Primary Investigator (PI) Name
Sumit Chakravarty
Integrating Wireless Mesh Networking and Object-Oriented Semantic Mapping into UAVs
Unmanned Aerial Vehicles (UAVs) are becoming increasingly utilized in applications such as disaster response, surveillance, and extending connectivity. However, reliance on human operators and centralized communication systems introduces risks of human errors, and miscommunication, which has led to serious accidents such as the aircraft collision in DC. Ensuring reliable communication and collision avoidance among UAVs is a major factor in operational success. This research explores how Wireless Mesh Networking (WMN), and Object-Oriented Semantic Mapping could be implemented with UAVs to improve wireless communication and collision avoidance. WMNs are a type of network topology which allows for scalable, decentralized communication with adaptive routing. Object-Oriented Semantic Mapping creates 3D maps providing object-specific data, improving environmental awareness. WMNs enhance UAVs by providing decentralized communication paired with robust connectivity and adaptive routing, allowing UAVs to effectively communicate and maintain connectivity in dynamic environments. Choosing routing and distributed MAC protocols such as AODV and CSMA/CA respectively or others, creates a robust communication network and improves efficiency. Object-Oriented Semantic Mapping enhances UAV navigation by providing object detection and probabilistic mapping, allowing for UAVs to interpret and respond to obstacles effectively. By integrating YOLOv8-based real-time object detection and probabilistic SLAM techniques, UAVs can create detailed maps that can identify obstacles and improve navigation decisions. The integration of these technologies allows for the creation of a UAV network capable of real-time collaboration, and obstacle avoidance. The adjustments prove Wireless Mesh Networking together with Object-Oriented Semantic Mapping enables UAVs to operate with better collision avoidance as well as robust communication. The combination of real-time object detection and probabilistic mapping with optimized data transmission enables UAVs to work better in dynamic conditions while requiring minimal human supervision. Future research aims to perfect these technologies because their applications in self-driving cars and aircraft designs contributes to more reliable autonomous systems.