As one of the most important issues in the field of mobile robotics, self-localization allows a mobile robot to identify and keep track of its own position and orientation as the robot moves through the environment. In this work, a hybrid localization approach based on the particle filter and particle swarm optimization algorithm is presented, focusing on the localization tasks when an a priori environment map is available. This results an accurate and robust particle filter based localization algorithm that is able to work in symmetrical environments. The performance of the proposed approach has been evaluated for indoor robot localization and compared with two benchmark algorithms. The experimental results show that the proposed method achieves robust and accurate positioning results in indoor environments, requiring fewer particles than the benchmark methods. This advance could be integrated in a wide range of mobile robot systems, helping to reduce the computational cost and improve the navigation efficiency.

Introduction

Along with the technological advancements in the field of mobile robotics, research interest in autonomous mobile robots has been increasing in the past decades. A diverse range of applications in rescue (Michael et al., 2014), mining (Ma & Mao, 2018), agriculture (Bengochea-Guevara, Conesa-Muñoz, Andújar, & Ribeiro, 2016), military (Miksik, Petyovsky, Zalud, & Jura, 2011) and civilian tasks (Choi, Lee, Viet, & Chung, 2017; Le, Phung, & Bouzerdoum, 2014; Song, Gao, Ding, Deng, & Chao, 2017) encourage researchers to carry out research works in mobile robotics. Self-localization is a prerequisite for successful deployment of an autonomous mobile robot since it identifies the robot’s pose (position and orientation) as it moves in the environment. By providing an “absolute” position estimate to the map frame, robot localization is one of the critical issues for mobile robot systems and it is typically the foundation of a variety of tasks, including map building, autonomous navigation, mobile manipulation, target tracking, etc. The mobile robot localization problem falls into two main categories: global localization (GL) and local pose tracking (relocalization) (Thrun, Burgard, & Fox, 2005). The local pose tracking problem assumes that the initial pose of the robot is already known, and it tries to keep track of the robot state over time. The GL problem is fundamentally different because no prior knowledge about the robot’s position is available, hence the robot has to locate itself from scratch and reduce the ambiguities of pose estimates.