Integration of Human-Driven and Autonomous Vehicle: A Cell Reservation Intersection Control Strategy

Abstract

With the advent of driverless automobiles, there is a tremendous opportunity to increase the effectiveness of the traffic system, user comfort, and reduce traffic accidents brought on by human mistake. It seems inevitable that driverless and human-driven vehicles will coexist. The difficulties in building new AV roads will be overcome by the coexistence of traffic, and this method builds on already-existing road infrastructure. Attempts to combine human and autonomous-driven cars have raised a crucial issue in road traffic management: What effects can be expected when a certain number of autonomous vehicles coexist peacefully with human-driven vehicles in terms of efficiency and safety? Two-dimensional (2D) lateral and longitudinal vehicle behavior defines traffic coexistence. It is adapted and improved upon the car-following model idea to effectively depict a mixed traffic system. For secure, slick, and effective mixed-traffic management, the “Cell Reservation-based Intersection Control Management Strategy” is suggested. The key contributions of this research include a guide to mixed traffic integration pattern, an extension of the existing 1-dimensional homogeneous car-following model strategies to a 2-dimensional heterogeneous traffic system, an improvement in human-driven vehicle performance when autonomous vehicle inter-vehicle distance is adjusted, and a method for harmonising speed in mixed traffic. A physics agent traffic simulator is created and used to test three traffic management strategies, including the traffic light method, the collision avoidance with safe distance method, and this proposed method, in order to determine the benefits of the cell reservation traffic control strategy. Experiments with various vehicle type ratios were done to validate the model. The collected findings show that the cell reserve method outperforms the alternatives in terms of performance gain.