Autonomous vehicles’ object detection architectures ranking based on multi-criteria decision-making techniques

Autonomous vehicles rely on object detection for safety, and various algorithms have been developed to improve speed and precision. This paper ranks 44 object detection algorithms published between 2019–2022 using COPRAS and VIKOR ranking techniques. 24 algorithms were built using LiDAR point clouds, with Convolutional Neural Networks (CNN) being the foundation. Second + MD3D, CenterPoint + MD3D, and 3D Point Cloud Object Detection Algorithm are the top three algorithms for 3D object detection using Copra’s methodologies. FlexiNet, YOLOv2-5D, and CSA-SS are the best 2D object detection algorithms. These methods are suitable for autonomous driving systems, as accuracy and speed are essential requirements. © The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of Computer Applications and Management 2023.