Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., Desmaison, A., Kopf, A., Yang, E., DeVito, Z., Raison, M., Tejani, A., Chilamkurthy, S., Steiner, B., Fang, L., Bai, J. and Chintala, S. (2019) PyTorch: An Imperative Style, High-Performance Deep Learning Library. In: Wallach, H., Larochelle, H., Beygelzimer, A., d’Alché-Buc, F., Fox, E. and Garnett, R., Eds., Advances in Neural Information Processing Systems, Neural Information Processing Systems Foundation Inc. (NeurIPS), Vancouver, 8024-8035.

has been cited by the following article:

• TITLE: Compare and Contrast LiDAR and Non-LiDAR Technology in an Autonomous Vehicle: Developing a Safety Framework

  AUTHORS: Benjamin Quito, Larbi Esmahi

  KEYWORDS: AI, ML, Image Classification, Safety Frameworks, LiDAR, Non-LiDAR

  JOURNAL NAME: Open Journal of Safety Science and Technology, Vol.13 No.3, September 4, 2023

  ABSTRACT: Ensuring safety has always been of utmost importance in vehicular operations, where traditionally, human drivers have been solely responsible for driving. However, with the emergence of advanced technologies, we are now on the brink of a new era with the introduction of Autonomous Vehicles (AVs), in which control over the vehicle gradually shifts to Artificial Intelligence (AI). The safety of these vehicles has raised concerns among the public. In terms of safety, human drivers heavily rely on visual perception. Key elements that contribute to safe driving include situational awareness, vehicle control, reaction capabilities, and anticipation of potential hazards. With the introduction of AVs, these fundamental factors remain unchanged. AVs rely on two primary technologies, namely LiDAR and Non-LiDAR, to perceive their surroundings. This research focuses on three primary aspects. Firstly, it involves the development of an image classification model to assess the safety of AVs. This model determines whether the images captured by LiDAR and Non-LiDAR technologies can be accurately predicted using supervised learning. An algorithm is employed to identify the input images obtained from both LiDAR and Non-LiDAR technologies. The results demonstrate that the model achieved a high accuracy rate of 94.63% in predicting the images. Secondly, a Safety Framework is established to facilitate the subsequent proposal for Experimental Research, which is the third aspect. The Safety Framework incorporates the application of LiDAR and Non-LiDAR technologies mounted on a vehicle that is operated in diverse weather conditions by both a human driver and an autonomous system. The Scenario tables, which are currently in a blank state, will be populated upon completion of the Experimental Research. The Experimental Research aims to compare and contrast the performance of LiDAR and Non-LiDAR technologies on a vehicle driven by a human operator versus an Autonomous Vehicle. The findings of this research will be documented in the Scenario table outlined within this study, ultimately shedding light on the safety implications of implementing LiDAR and Non-LiDAR technologies within an AV context.