Vol.:(0123456789) 1 3 Iran Journal of Computer Science https://doi.org/10.1007/s42044-020-00060-4 ORIGINAL ARTICLE ANFIS controller design based on pigeon‑inspired optimization to control an UAV trajectory tracking task Boumediene Selma 1  · Samira Chouraqui 1  · Belkacem Selma 2  · Hassane Abouaïssa 3 Received: 10 December 2019 / Accepted: 2 May 2020 © Springer Nature Switzerland AG 2020 Abstract Unmanned aerial vehicles (UAVs) are fying platforms that have become increasingly used in a wide range of applications. However, the most recent research has aimed to improve the quality of UAV control to achieve its mission accurately. This paper proposes a robust and intelligent control method based on adaptive neuro-fuzzy inference system (ANFIS), and pigeon- inspired optimization algorithm (PIO) to govern the behavior of a three-degree of freedom (3-DOF) quadrotor UAV. The quadrotor was chosen due to its simple mechanical structure; nevertheless, these types of UAVs are highly nonlinear. Intel- ligent control that uses artifcial intelligence computing approach such as fuzzy logic is a suitable choice to better control nonlinear systems. The ANFIS controller is proposed to control the movement of UAV to track a given reference trajectory in 2D vertical plane. The PIO is used to obtain the ANFIS optimal parameters with the aim of improving the quality of the controller and therefore, to minimize tracking error. To evaluate the performance of the ANFIS controller tuned by PIO, a comparison between the proposed ANFIS-PIO, ANFIS and proportional–integral–derivative controllers is illustrated, and comparative results demonstrate that the proposed controller is more efective. Keywords Unmanned aerial vehicle (UAV) · Robust control · Adaptive neuro-fuzzy inference system (ANFIS) · Pigeon- inspired optimization (PIO) 1 Introduction Unmanned aerial vehicle (UAV), or commonly called drone, is an aircraft without human guidance. Either UAV can be operated with a remote control and also can fly autonomously according to pre-programmed plans or other more complex dynamic systems. The UAV technology has been continuously evolving with exceptional growth over the last years [1], leading to the emergence of a large number of services ofered and potential applications. Drones are not meant to only serve military purposes [2], but have also become widely used in civilian and industrial domain, such as logistics and transportation [3, 4], photography and flm- making [5], safety and security [6], mapping [7], agriculture [8, 9], monitoring [10, 11], surveillance [12], architecture [13, 14] and many other applications. Its use is increasing in most areas due to their low maintenance cost, ease of deployment, high mobility and hovering ability [15–17]. The overall purpose of this study is to develop a robust and intelligent control method for nonlinear systems classes, without precise knowledge of system in question, providing system stability and robustness. The introduction of expert knowledge into the control system perform the task in the same way as a human operator, able to able to keep the sys- tem under control without knowing the mathematical model. This approach is often referred to intelligent control, and can be supplemented by a level of supervision that allows * Boumediene Selma selma.boumediene@yahoo.fr Samira Chouraqui s_chouraqui@yahoo.fr Belkacem Selma selma.belkacem@yahoo.com Hassane Abouaïssa hassane.abouaissa@univ-artois.fr 1 Département d’Informatique, Université des Sciences et de la Technologie d’Oran USTO’MB, 31000 Oran, Algeria 2 Département de Génie Electrique, Faculté des Sciences et de la Technologie, Université de Mostaganem, 27000 Mostaganem, Algeria 3 Laboratoire de Génie Informatique et d’Automatique de l’Artois (LGI2A), Univ. Artois, EA 3926, 62400 Béthune, France