Path planning for drone swarms

• Main Author: Lin, Zhi
• Other Authors: Mir Feroskhan
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering
• Online Access: https://hdl.handle.net/10356/177439
• Institution: Nanyang Technological University

Unmanned Aerial Vehicles (UAV) drones has been undergoing major improvements and developments, and has developed significant popularity over the years. With this development, comes different types of applications it can access to. However, there are still research gaps that needs to be covered. Hence, the Singapore Amazing Flying Machine Competition, or SAFMC for short[1], was organized, to invite students of all ages, to participate and develop innovative algorithms and solutions. For this competition, it simulates a Search and Rescue Mission, where the objective is to develop a multiagent system, which a swarm of drones will be deployed and fly into an indoor environment autonomously, to find the victims trapped in the building. The approach of this project can be summarized into three parts. First, a path planning algorithm will be created to set up the autonomous nature of the multiagent system, which the algorithm will be based on known algorithms, such as the Pigeon-Inspired Optimization Algorithm (PIOA) and the Bacteria Foraging Optimization Algorithm (BFOA). Second, a decentralized Ground Control Station (GCS) will be developed, to control the swarm of drones simultaneously, without communication interferences between the drones. Lastly, in-built sensors such as the bottom facing camera, will be used not only to localise the positioning of the drones, but to also detect the victims on the ground. Taking this approach, the multiagent system was able to successfully complete its objective outcome, to a certain degree, of finding most of the victims trapped in the building. Moving forward, to continue the research and development for this project, certain implications and recommendations have to be considered. Some future works to look into are, the limitations on the in-built sensors, researching on other sensors with a higher degree of accuracy, and create a GCS, that is not only decentralized, but also allow the establishment of communications between the drones within a swarm system.