Autonomous Mobile Robot Navigation and Obstacle Avoidance: A Comprehensive Review

Abstract

The rise of AMRs has changed our perception and also our interaction with automation. At the centre of this transformation are navigation and also obstacle avoidance, both equally critical requirements for deploying AMRs in various settings. This thorough review examines the front-line progress in navigation and collision avoidance for AMRs, touching upon numerous contemporary approaches and methodologies algorithms along with technologies that seek to enhance functionality. The paper offers a detailed review of the established approaches, such as rule-based approaches; potential fields; reactive navigation systems as behaviour systems and path following algorithms that have been amassed to face the challenge in practice. Conversely, technological developments in terms of machine learning, computer vision sensor fusion and SLAM algorithms as well as edge computing are discussed in the aspect that they have an unprecedented impact on AMR navigation. Global and local approaches are approached through universal international optics and also national adjustments that reveal the peculiarities of separate countries. The Data Analysis and Processing section points out the importance of technologies, which define AMR performance. Due to the constraints obtained from previous studies, it is very clear that additional research has to be done in relation to concentrating on fixing gaps under the controlled environments and using standard benchmarks; sensor heterogeneity issues; and also practical implementation of theoretical aspects. Briefly speaking, this review serves as a map for the intricate landscape of AMR navigation and also obstacle avoidance. Its main goal is to support the ongoing debate, stimulate innovation and identify new research directions in a rapidly changing world of autonomous mobile robotics that breaches the existing deployment barriers.