Chicken Swarm–Based Method for Ascent Trajectory Optimization of Hypersonic Vehicles
Yongtao Li,Yu Wu,Xiangju Qu
DOI: https://doi.org/10.1061/(asce)as.1943-5525.0000757
IF: 2.242
2017-01-01
Journal of Aerospace Engineering
Abstract:Trajectory optimization of hypersonic vehicles has become a hotspot because of its advantages in flight speed and range. This task is challenging because of the strong nonlinear couplings among the aerodynamics, propulsion, and flight state of hypersonic vehicles. In this paper, a novel chicken swarm-based method is proposed to solve the ascent trajectory optimization problem. Firstly, the ascent trajectory optimization problem is formulated into the optimal control problem, and the cost function that ensures the minimum fuel consumption, which is subjected to various constraints such as dynamic pressure, load factor, and aerodynamic heating, is proposed. Then the principles of the chicken-swarm optimization (CSO) algorithm are explained, and the CSO-based trajectory optimization method is proposed. With the method, the control variables are discretized at a set of Chebyshev collocation points, and these points are the variables to be optimized in the CSO algorithm. Also, a segmented strategy is introduced to treat the constraints discriminately in the penalty terms of the cost function according to the flight state of the hypersonic vehicle. Based on such a strategy, the process of trajectory optimization with the CSO algorithm is depicted. A series of comparative experiments were conducted to investigate the feasibility and superiority of the proposed method. Furthermore, more experimental results are presented to discuss the influence of hierarchical order and collocation point selection on the performance of the CSO-based method. (C) 2017 American Society of Civil Engineers.