Executive Summary : | Forest fires are a significant emergency that cause significant damage to natural and urban areas. In India, such as Himachal Pradesh, the detection and response time of mitigating fires is crucial. Robotic technology, such as a swarm of drones, can help mitigate these fires safer and faster. However, drone reliability is crucial, as undesired actions may lead to additional damage. This project proposes an experimental framework to handle faulty situations when drones work collaboratively to mitigate forest fires. The time-varying formation tracking algorithm (TVFT) is proposed, where an arbitrary formation can be decided online. Any anomaly in the swarm is diagnosed and detected online by a distributed observer at each drone. The isolation of faulty agents is determined based on fault magnitude, and if isolated, proper control inputs are given to ensure safe landing. The project focuses on a practical approach with only input and output data available at a given time. The Koopman framework is used to investigate fault-tolerant control of a swarm of drones, implementing the TVFT, FTC, and reformation of the swarm. This experimental framework strengthens and supports the implementation of a swarm of drones in real-time applications and encourages researchers to explore the Koopman framework and its idea while working with nonlinear control systems. |