Hybrid Self-controlled vehicles for averting head-on collision

Abstract

The excessive cases of fatal vehicle collisions are major concerns across the globe. Nevertheless, there are insufficient empirical simulations to illustrate the stages involved in head-on collisions between a smart vehicle and another smart vehicle over the years. Studies have also shown that about 90% of instant death recorded on most roads is caused by head-on collisions between two vehicles. Though vehicle collisions can occur between a vehicle and other objects such as tree(s), stumbling block, animal(s), etc, however, significant numbers of training schools do not have access to empirical software products that drivers can use to learn how they can lessen the impacts of head-on collision. Thus, most victims often suffer severe loss such as stern injury, permanent disability, death, when head-on collision in accidentally happen to them. Consequently, several road safety measures have been implemented in recent decades but they have limited success till date. Thus, this paper explores the above issues and proposes self-controlled vehicles to prevent them. The self-controlled vehicles are controlled by four different categories of inbuilt rules that have capabilities to instruct and regulate two vehicles commuting in opposing directions and they are implemented with Python programming language and relevant libraries. Series of evaluations with Gini impurity suggest the distributions of four basic stages before head-on collisions would occur between two vehicles. The results also suggest that there are most excellent distances of separation that two opposing vehicles must take the decisions that would enable them avoid head-on collisions. Finally, the designs can be valuable tools to driving schools in teaching and counseling prospective drivers on road accidents before issuance of driving license to them.