Human-Robot Interaction and Self-Driving Cars Safety Integration of Dispositif Networks

Abstract

In recent years, despite many researches and progress in artificial intelligence, we have witnessed many accidents involving self-driving cars. For self-driving cars to have potential for positive impact on road safety, a different Human-Robot Interaction (HRI) model is required that provides a learning algorithm mechanism to recognize other vehicles, not just as a moving object, but as a vehicle intelligently controlled by a human driver. Then, self-driving cars may successfully deliver on their promise to save thousands of lives annually. Current algorithms used in the development of self-driving cars are mainly invested in the deep learning of which neural networks need to be trained on representative datasets that include examples of all possible driving, weather, and situational conditions. Until recently, HRI was researched in light of human perception of self-driving cars and improved collision avoidance, and to predict other driver's intentions based on monitoring their movement. However, with recent accidents involving self-driving cars, more than at any other time, there is a need for an advanced HRI model to improve safety and human trust for autonomous vehicles. A human driver's way of thinking leads them to make certain decisions which may not be logical or familiar to current robot algorithms. For humans, factors shaping the way of seeing and behaving are not static; rather, they are varying in different societies, cultures, and countries, and are also subject to continuous changes. Such factors are explained by researchers as inter-related networks of dispositifs. In this paper, we present, that if self-driving cars are able to integrate such dispositifs networks within their HRI model, by creating two algorithms; a) local-signature; and, (b) individual-signature for regional and on-road; then, it would be more likely and globally possible to predict what humans will do on the road, thereby correctly determining how to behave appropriately around them.