Novel two phase algorithm to design three loop autopilot using parameter plane technique and particle swarm optimization

详细信息    查看全文

• 作者：Siddhardha Kedarisetty ^{siddhardhak2010@gmail.com" class="auth_mail" title="E-mail the corresponding author}
• 关键词：Lateral acceleration autopilot ; Parameter plane technique ; Particle swarm optimization
• 刊名：IFAC-PapersOnLine
• 出版年：2016
• 出版时间：2016
• 年：2016
• 卷：49
• 期：1
• 页码：830-835
• 全文大小：677 K

文摘

Three loop autopilot is a popular autopilot structure used in aerial vehicles. Design of autopilot is based on user requirements that are given in terms of time domain specifications (like rise time, maximum overshoot, etc.) and frequency domain specifications (like gain and phase margins). In literature, it has been shown that three loop lateral acceleration autopilot for a rigid tactical aerospace vehicle (TAV) can be designed using time domain techniques such as pole placement or frequency domain techniques such as parameter plane technique. However an autopilot designed using time domain techniques may not satisfy frequency domain specifications and vice versa. This paper proposes a two phase algorithm for three loop autopilot design for a TAV such that both the time domain and the frequency domain requirements are met. In the first phase, the space of the control gains of the three loop autopilot is reduced to a subspace from which if control gains are chosen, the desired stability margins are guaranteed. This phase uses parameter plane methodology. The subspace so obtained is used in the second phase as a domain on which particle swarm optimization is used to obtain the optimal set of control gains that minimizes the deviation of the system from required time domain specifications. The efficacy of the proposed algorithm is illustrated through three loop lateral acceleration autopilot design example for a rigid body TAV.