About

Efficiency and Safety collaborative traffic functions in smart intersections

What?

The problems of increased traffic, consumption, and pollution, decreased safety or driving stress, became very stringent nowadays, as they affect our daily lives. There are existing solutions for some of the involved sub-problems, but none of them tackles the whole issue in a systematic and integrated manner. Our proposed solution is based on creating the ability of passenger cars to coordinate their driving behavior when passing through intelligent intersections equipped with smart infrastructure.

How?

The coordination is based on internal vehicle intelligence (related to environment perception and self-driving capabilities) and communication with other vehicles as well as with the smart infrastructure based on connectivity technologies as 5G. To enable coordinated driving behavior, Swarm Intelligence (SI) technology builds up onto the Automated Driving (AD) framework. Within this framework, shared collaborative decision rules and coordination of movements enhance the ability to withstand external disturbances and measurement errors.

The smart infrastructure senses the upcoming vehicles with environment sensors (vision and/or radars) and short-range communication. It builds a comprehensive environmental model (EM) and the generated list of objects is shared via specific wireless communication to the vehicles. The vehicle – infrastructure cooperation is not exploited enough nowadays; thus, the unawareness of other traffic participants’ trajectories in intersections can lead to unoptimized path planning. As such, a significant benefit would be here for automated driving (AD) vehicles. Moreover, concepts of Swarm Intelligence can be incorporated in order to improve the ego vehicle path planning by optimizing a general goal related to traffic flow at the network level, through coordinated control actions. This can be even further improved by incorporating information received from nearby vehicles. Traffic and driving simulators (e.g., SUMO, CARLA, SUMMIT, VISSIM), based on synthetic and collected data will pre-validate the proposed solutions.

But not only the AD vehicles can benefit from the knowledge of Intersection EM. Safety of the road users inspires the Vision Zero (zero accidents, zero fatalities) for all stakeholders, be it manufacturers or road authorities. The EM information can be used to protect road users (especially vulnerable road users like pedestrians, bicycle, or e-scooter riders) in possibly dangerous situations, like occluded incoming traffic.

In sum, this Use Case consists of the implementation of traffic management features using multi-access edge computing (MEC) capabilities to enable low-latency V2X communication, divided into two categories, one supporting traffic efficiency and the other one traffic safety.

5G technology

The characteristics of 5G technology will definitely help in realizing the scope of the project:

  • The very precise positioning and ranging offered by the 5G Positioning and Location services supports cooperative and automated driving.
  • The ultra-low latency and ultra-high reliability communication between vehicles is useful to support high-density vehicle grouping and to improve the grouped driving by inter-vehicle cooperation. Information vehiculated refers mainly to driving parameters.
  • At the same time, the ultra-high throughput and ultra-low latency connectivity would enable the exchange of raw or processed data gathered through local sensors and/or live video images that can be employed to improve safety by video analytics, e.g., other cars abnormal behavior.
  • The ultra-high throughput helps to build local, dynamic maps based on camera and sensor data. This data can be distributed at street intersections to exchange information with cars and local authorities. For example, connectivity could be used to supply a driver or an on-board computer with a bird’s eye view of an intersection or see-through capability when driving behind a truck.
  • Moreover, the 5G technology will be able to support a very large number of simultaneous connections in a small geographic area, enabling each vehicle to gather more information about its immediate surroundings.

 

The contribution from Orange at the overall project reference architecture should be seen mainly from the perspective of a Service Provider (end user service chaining) and full ecosystem Infrastructure Provider (networking and computing capabilities) related to 5G infrastructure.

Project Phases

The project can be divided in two phases:

  1. Initial development of algorithms for: (i) dynamics control based on swarm-defined collaborative actions at individual/group vehicle level, (ii) object detection and localization of the road users, and comprehensive environment model creation at the intelligent intersection level – TUIASI; (iii) precise localization of the road users, based on radio technologies – TUIASI;
  2. Real-time implementation and testing of the algorithms developed in the first phase by using the available 5G infrastructure, which should yield in a validated architecture – TUIASI & Orange; for example, the developed algorithms could be tested in Podu Ros area in which Continental plans to install various sensors for environment perception and comprehensive model creation.

Our Team

Our team includes several professors, experienced researchers, PhD and master students from 2 faculties of Gheorghe Asachi Technical University of Iasi, Romania: Faculty of Automatic Control and Computer Engineering and Faculty of Electronics, Telecommunications
and Information Technology.

Co-founder INTERACT
(TUIASI & Continental)
Co-founder INTERACT
(TUIASI & Continental)

Member of INTERACT (TUIASI)

Member of INTERACT (TUIASI)

Member of INTERACT (TUIASI)

Member of INTERACT (TUIASI)

Member of INTERACT (TUIASI)

Member of INTERACT (TUIASI)

Member of INTERACT (TUIASI)

Member of INTERACT (TUIASI)

Member of INTERACT (TUIASI)

View our solution

Our solution is based on creating the ability of passenger cars to coordinate their driving behavior when passing through intelligent intersections equipped with smart infrastructure.