R ecently, quadrotors gained popularity due to the- ir high maneuverability, cost and vertical take- off/landing capabilities. Nevertheless, they also have disadvantages such as they have a limited flight time and small payload capabilities. In addition to these, a major part of the energy of a quadrotor is spent aga- inst gravity for hovering. Still, they are one of the most adopted air vehicles for commercial and research purposes. Most of the time UAV’s (Unmanned Aeri- al Vehicles) are used in outdoor applications such as surveillance, search/rescue and patrolling. Recently, UAV’s started to find uses in indoor environments. Material handling in manufacturing and inspection in harsh environments are to name a few[1] of these applications. One of the most promising applications is to utilize them in urban relief and disaster operati- ons where a UAV moves autonomously avoiding obs- tacles in GPS-denied buildings to help human opera- tors for rescue operations. In order to navigate in an indoor environment, a map of the environment is needed. However, in most of the cases either the map is unknown or some partial information about the environment is available. Indo- or mapping process can be performed by using seve- Article History: Received: 2020/03/04 Hittite Journal of Science and Engineering, 2020, 7 (2) 125-134 ISSN NUMBER: 2148-4171 DOI: 10.17350/HJSE19030000181 A Novel Navigation Algorithm for Mapping Indoor Environments with a Quadrotor Omer Oral 1 Ali Emre Turgut 1 Kutluk Bilge Arikan 2 1 Middle East Technical University, Department of Mechanical Engineering, Ankara, Turkey 2 TED University, Department of Mechanical Engineering, Ankara, Turkey Accepted: 2020/05/22 Online: 2020/06/26 Correspondence to: Omer Oral, Middle East Technical University, Department of Mechanical Engineering, 06800, Ankara, TURKEY E-mail: omroral@gmail.com ral methods. SLAM technique is usually employed for mapping. SLAM (Simultaneous Localization and Map- ping) is the process of simultaneous map extraction and robot localization. The difficulty of this process is inherent in its definition. A map is required for correct localization, while an accurate localization is required for mapping [2]. In indoor environments, ground-based robots were commonly employed for map extraction [3]. In studies with ground robots, while LIDAR (Laser Imaging Detection and Ranging) and IMU (Inertial Measurement Unit) were mostly used, Omara et al. [4] used a Kinect sensor for exploration and mapping of the environments using SLAM methods. Different sensors were utilized when UAV’s were used for mapping. John son [5] used IMU, optic flow sensors and a camera to navigate autonomously in indoor environments. Due to unreliable attitude estimations with inertial sensors, Hough transform was adopted for attitude estimation of the quadrotor. Ahrens et al. [6] also used a quadrotor equipped with IMU and a camera to navigate in indoor environments with obstacles. They used ”good features to track” detector for feature tracking and use these features for navigation and mapping. Roberts et al. [7] developed a quadrotor capable of navigating in indoor environments autonomously using and IMU, an ultra- ABSTRACT I n the last decade, unmanned aerial vehicle gained popularity and started to be used in different tasks most of which are performed in outdoor environments. Still, there is a great potential to use quadrotors in indoor tasks such as urban relief and disaster operations. In this paper, we developed a framework and a novel target-based navigation algorithm for mapping of an unknown 2D environment with a quadrotor using an ultra- wideband system. The target-based navigation algorithm aims to explore map of the en- vironment by moving the border between the discovered and undiscovered areas. It uses A* search algorithm for path planning if there is an obstacle present in the environment. The target-based navigation algorithm is implemented on Gazebo simulator and its perfor- mance is compared with the well-known wall following algorithm and exploration algo- rithm in terms of task completion time and distance travelled. The target-based navigation algorithm outperforms the other two algorithms especially in environments with obstacles. INTRODUCTION Keywords: LIDAR; SLAM; Mapping; UWB; Trilateration technique; Quadrotor; UAV; Indoor; Obstacle avoidance.