Identification for a Reentry Vehicle Via Levy Flight-Based Pigeon-Inspired Optimization

Daifeng Zhang,Haibin Duan
DOI: https://doi.org/10.1177/0954410016682274
2016-01-01
Proceedings of the Institution of Mechanical Engineers Part G Journal of Aerospace Engineering
Abstract:Reentry vehicles are valuable both in the military domain and scientific research. However, the high-order nonlinearity of mathematical models is always a bottleneck for reentry study. One of the reasons is the difficulty to measure the involved aerodynamic parameters. Therefore, parameter identification is a crucial issue in modeling and controller design for reentry vehicles. This paper mainly focuses on the identification approach for aerodynamic parameters of an existing reentry vehicle. Wind field turbulences are modeled to imitate the real flight scenarios. A novel bio-inspired optimization algorithm is proposed to solve this problem. Our proposed method stems from the pigeon-inspired optimization, which is an effective swarm intelligence optimizer utilized in many research areas. Typical characteristics of Levy flight are drawn on to improve the global accuracy of the new algorithm. Finally, comparative experiments with some homogenous methods are conducted to verify the expected performances of our identification algorithm.
What problem does this paper attempt to address?