Login / Signup

Effective and Safe Trajectory Planning for an Autonomous UAV Using a Decomposition-Coordination Method.

Imane NizarAdil JaafarZineb HidilaMohamed BarkiEl Hossein IlloussamenMohammed Mestari
Published in: Journal of intelligent & robotic systems (2021)
In this paper, we present a Decomposition Coordination (DC) method applied to solve the problem of safe trajectory planning for autonomous Unmanned Aerial Vehicle (UAV) in a dynamic environment. The purpose of this study is to make the UAV more reactive in the environment and ensure the safety and optimality of the computed trajectory. In this implementation, we begin by selecting a dynamic model of a fixed-arms quadrotor UAV. Then, we define our multi-objective optimization problem, which we convert afterward into a scalar optimization problem (SOP). The SOP is subdivided after that into smaller sub-problems, which will be treated in parallel and in a reasonable time. The DC principle employed in our method allows us to treat non-linearity at the local level. The coordination between the two levels is achieved after that through the Lagrange multipliers. Making use of the DC method, we can compute the optimal trajectory from the UAV's current position to a final target practically in real-time. In this approach, we suppose that the environment is totally supervised by a Ground Control Unit (GCU). To ensure the safety of the trajectory, we consider a wireless communication network over which the UAV may communicate with the GCU and get the necessary information about environmental changes, allowing for successful collision avoidance during the flight until the intended goal is safely attained. The analysis of the DC algorithm's stability and convergence, as well as the simulation results, are provided to demonstrate the advantages of our method and validate its potential.
Keyphrases
  • dendritic cells
  • machine learning
  • mental health
  • healthcare
  • primary care
  • immune response
  • computed tomography
  • risk assessment
  • human health
  • network analysis