Application of Machine Learning in Ethical Design of Autonomous Driving Crash Algorithms

Abstract

The age of algorithms is here, and it is really changing people’s lives. More and more ethical problems related to algorithms have attracted people’s attention, but the related ethical research is still far behind the research of algorithms. As more intelligent algorithms emerge in an endless stream, there will also be a lot of algorithmic ethical issues. On the other hand, with the continuous improvement of the development level of the automobile industry, people have a stronger demand for the safety and stability of modern transportation, and more and more autonomous driving technology has been promoted and applied in the market. At present, most of the studies on the longitudinal collision avoidance system of vehicles use collision warning or emergency braking to avoid collision. However, when the vehicle is in a special situation such as high speed and slippery road, emergency steering is more effective. In order to further improve the vehicle safety and ethical algorithm design points, this article revolves around vehicle lateral active collision avoidance control method research, the collision avoidance decision-making, and path planning and collision avoidance transverse vehicle longitudinal motion control is analyzed, and based on automated driving simulation experiment, the tests carried out to verify the designed control strategy. The experimental results show that the proposed method not only has a good effect of preventing automatic driving collision but also can meet the requirements of algorithm ethics. This research can effectively guide the research of algorithmic ethics in the field of autonomous driving and effectively reduce the occurrence of traffic accidents.

Advancing autonomy through lifelong learning: a survey of autonomous intelligent systems

“…The development of the automobile industry has driven an increased demand for safety and stability in modern transportation. As a result, autonomous driving technology has gained significant traction and is being widely deployed in the market (Xiao, 2022). This technology is revolutionizing intelligent transportation and smart city systems by enhancing the efficiency and safety of transportation networks.…”

SDC-Net: End-to-End Multitask Self-Driving Car Camera Cocoon IoT-Based System

“…Recently, deep learning and computer vision are invading the automotive field. Sophisticated features have become essential in order to achieve a self-driving car, including: lane keeping assist (LKA) [ 1 ], path planning [ 2 ], adaptive cruise control (ACC) [ 3 ], automatic emergency braking (AEB) [ 4 ], traffic jam assist (TJA) [ 5 ], and crash avoidance (CA) [ 6 ]. AEB aims to force a brake control especially in case of an object in front of the ego-vehicle, LKA aims to control the ego-vehicle steering angle within the same lane with at constant speed.…”

Deep Learning and Autonomous Vehicles: Strategic Themes, Applications, and Research Agenda Using SciMAT and Content-Centric Analysis, a Systematic Review

“…How challenges be mitigated to ensure responsible deployment of AI in AVs? [80][81][82] Transdisciplinary themes in DL-AV…”

“…Privacy concerns related to data collection and usage must be addressed to protect the personal information of users and maintain their trust. Moreover, considering the societal impact of AVs is crucial to ensure that they align with ethical principles and contribute positively to communities, fostering equitable access and social well-being [80][81][82].…”