IAES International Journal of Artificial Intelligence (IJ-AI) Vol. 12, No. 3, September 2023, pp. 1033~1043 ISSN: 2252-8938, DOI: 10.11591/ijai.v12.i3.pp1033-1043  1033 Journal homepage: http://ijai.iaescore.com A review on object detection for autonomous mobile robot Syamimi Abdul-Khalil 1 , Shuzlina Abdul-Rahman 1,2 , Sofianita Mutalib 1,2 , Saidatul Izyanie Kamarudin 1 , Siti Sakira Kamaruddin 3 1 School of Computing Sciences, College of Computing, Informatics and Media, Universiti Teknologi MARA, Selangor, Malaysia 2 Research Initiative Group of Intelligent Systems, College of Computing, Informatics and Media, Universiti Teknologi MARA, Selangor, Malaysia 3 School of Computing, Universiti Utara Malaysia, Kedah Darul Aman, Malaysia Article Info ABSTRACT Article history: Received Jan 17, 2022 Revised Dec 29, 2022 Accepted Jan 10, 2023 The advancement of autonomous mobile robots (AMR) is vastly being discovered and applied to several industries. AMR contributes to the development of artificial intelligence (AI), which focuses on the growth of human-interaction systems. However, it is safe to understand that mobile robots work closely in real-time and under changing surroundings. Similarly, some limitations may affect the efficiency of mobile robots. Thus, to improve the system's efficiency and accuracy, mobile robots should adopt the ability to detect incoming obstacles accurately. This paper presents the findings of a brief technology review aimed at identifying the current state of the art and future needs for AMR in object detection. This review paper is presented in the form of a narrative-literature review. Review articles were collected from 2015 until 2022 from journals or conference papers from well-known sources like IEEE Xplore, Science Direct, Scopus, and Web of Science (WOS). The analysis of the articles was discussed in four main topics, AI, object detection, AMR, and deep learning. Keywords: Artificial intelligence Autonomous mobile robot Deep learning Object detection This is an open access article under the CC BY-SA license. Corresponding Author: Shuzlina Abdul-Rahman Research Initiative Group of Intelligent Systems, College of Computing, Informatics and Media Universiti Teknologi MARA Selangor, Malaysia Email: shuzlina@uitm.edu.my 1. INTRODUCTION An autonomous mobile robot (AMR) is a system that operates in an unpredictable and partially unknown environment. AMR should have unrestricted movements and avoid any incoming obstacles within a surrounding [1]. Recently, AMR has been the core of technological advancement in daily services such as humanitarian assistance. It has been used as an autonomous agent in automotive, agriculture, education, and healthcare [2]. Hence, to accomplish intelligent systems, mobile robots are acquired to work remotely with computer systems, such as moving machine parts within a factory for storage or driving automated vehicles [3]. One of the challenges for the mobile robot is visual perception alongside interaction in the real world [4]. Object detection is a crucial robot vision system that trains them to perform complex tasks and overcome prevailing complications. For instance, one of robotics' primary duties is grasp detection, which helps robots collect objects in front of them [5]. Besides that, the AMR should provide the capability to detect any dynamic obstacles in real-time [6]. To make the robot function properly, the observation and integration using the navigation system, localization systems, and detection systems (sensors), along with motion and kinematics and dynamics systems are essential [7]. This paper will focus on reviewing the states-