Safety and Efficiency of Intersections With Mix of Connected and Non-Connected Vehicles

Abstract

Connected and autonomous vehicles have been significantly studied. They are connected to a network and communicate by exchanging information with each other, so they can detect blind spots that cannot be recognized by non-connected (conventional) vehicles. Therefore, they are expected to contribute to traffic efficiency and safety. However, even if connected vehicles are put to practical use in the future, it will take time to spread to the market, so it is considered that connected vehicles and non-connected vehicles will be mixed on the road. We proposed a method of enabling connected vehicles to share the information gathered from their sensors on surrounding vehicles near intersection in the mixed situation. We then examined the safety and efficiency of passing through an intersection through simulation. We found that efficiency improved and safety could be ensured compared to using conventional methods such as stopping before the intersection without using communication and using traffic lights. INDEX TERMS Connected vehicle, cooperative autonomous driving, V2V communication, mixed traffic.

A Survey on the Use of Container Technologies in Autonomous Driving and the Case of BeIntelli

“…Container architectures are therefore more scalable [31], [36], [37]. Typically, connected vehicles can communicate with each other within a limited range, e.g., 250 meters [38]. This provides little time window for message exchange and coordination among vehicles moving at high speeds.…”

Optimised Traffic Light Management Through Reinforcement Learning: Traffic State Agnostic Agent vs. Holistic Agent With Current V2I Traffic State Knowledge

“…The high relevance of the traffic control problem and the proven ability of RL to solve complex control problems led to a broad range of publications that combine the two topics [44]. The emergence of connected automated vehicles has led to a relatively new research field of traffic control for connected automated vehicles, see e.g., [45]- [50]. Our study focuses on conventional (non-automated) vehicles.…”

A Fundamental Study of Reliable Vehicle-to-Cloud Communication Using Multiple Paths with Redundancy Mitigation

“…The integration of sensing information and the cloud via V2X (vehicle to everything) is important for enabling automated driving and maneuver coordination [1][2][3][4]. One of the basic communication types in V2X is V2C (vehicle to cloud), which supports vehicles in updating and exchanging information on vehicle position, speed, and the results of object sensing.…”