DBCAR

Decisions and Behaviors for Cognitive Automobiles Research

 Partecipanti

Mappa

 Word cloud

Esplora la "nuvola delle parole (Word Cloud) per avere un'idea di massima del progetto.

 Obiettivo del progetto (Objective)

'The research presented herein belongs to the Intelligent Transport Systems and Services (ITS) field. ITS are application of information and communication technologies to the transportation system. The purposed work is to design behaviors and decisions for Autonomous Vehicles (vehicles that drive by itself, also known as the 'driverless cars'). After a correct interpretation of a traffic situation, the most appropriate behavior will be triggered, a path to follow will be designed, and the proper orders will be sent to the car actuators (steering wheel, throttle and brake).

The host center, “Institut für Mess- und Regelungstechnik” of the Karlsruher Institute of Technologie (KIT) conducts research on Cognitive Automobiles. Specifically it operates 2 experimental vehicles. One of them participated in the Darpa Urban Challenge competition of autonomous vehicles and successfully reached its finals. And the team at KIT has won the 2011 Grand Cooperative Driving Challenge, an international competition of cooperative vehicles which took place in the Netherlands.

Cognitive Automobiles research project is divided in three sub-projects: sub-project A: “Distributed Sensorial Perception”, sub-project B: “Behavior Decision, Generation, and Conduction”, and sub-project C: “Automobiles and IT Basis System”. This proposal will collaborate in sub-project B area. It will focus in two main tasks: First, the trajectories of cooperative vehicles at intersections are preplanned to optimize the expected flow of vehicles at a high safety level. Second, trajectories and maneuvers of cooperative vehicles are continuously monitored with respect to safety and are dynamically adjusted with respect to the actual safety level. To perform these cooperative maneuvers several vehicles will have to plan and negotiate trajectories very quickly.'

Introduzione (Teaser)

An EU-funded team developed software to control and coordinate autonomous vehicles. The modules allow groups of vehicles to communicate with a central station, facilitating the vehicles' smooth flow in traffic and maintenance of safe distances.

Descrizione progetto (Article)

Driverless vehicles, also known as autonomous vehicles, have been proposed as a solution to road congestion and safety issues. Such vehicles first need effective software control and decision-making systems.

The EU-funded 'Decisions and behaviors for cognitive automobiles research' (DBCAR) project aimed to develop a system allowing groups of several vehicles to coordinate their trajectories. Such coordination was intended to maintain safe distances between vehicles, and to improve the flow of vehicles in traffic. The project ran between May 2012 and May 2014.

Team members began with an extensive literature review concerning the coordination of the vehicles trajectories. As a result, the researchers chose the solution whereby autonomous vehicles are coordinated from a central agent. Hence, they developed the coordination software for the central agent and two software modules to interact with the control software of each vehicle. One module collected information from the vehicle database and relayed it to the central agent; the other receive the processed trajectories from the central agent and overcame the normal outputs of the vehicle controller.

Using information from all autonomous vehicles in a certain area, DBCAR developed cooperative trajectories for each vehicle, which were sent to the vehicles. Testing was initially conducted for cooperation of two vehicles, then modified to accommodate any number.

The research concluded that exploring all possible combinations is not feasible, and that alternative methods for optimal searching should be used instead. Testing showed the particle swarm optimisation method to be applicable, and the method can be expanded relatively simply to control numerous vehicles.

DBCAR produced methods for simultaneously controlling groups of vehicles through intersections and around other obstacles, increasing safety by maintaining a minimum safe distance. The broader outcome was methods for achieving fuel savings and reduction in travel time.