Search

Browse

• Communities
• Authors
• Year
• Titles
• Subjects

My Space

• Login

Movement strategies for emergency vehicles in lane-free environment

Pavlaki Ismini

Full record

---

Summary

Road congestion and road accidents persist as major problems, even after decades of research and developments in road safety and traffic management and call for comprehensive solutions. Automated vehicles receive high-quality information in a fraction of a second through an array of sensors that scan the surroundings frequently. With high-quality information and split-second decisions, automated vehicles have the potential to deliver safe and efficient traffic flow. Road traffic lanes were introduced to help human drivers for safe navigation at the cost of reduced utilization of the road space. Moreover, lane changing is a high-risk task, which has the potential to trigger congestion and possibly lead to road capacity drop. Full automation of traffic makes the lanes unnecessary and calls for the novel idea of lane-free traffic. In a futuristic scenario of 100% automated vehicles, there is no need for the vehicles to follow the rules designed for human drivers. A lane-free environment is considered for the current work with fully automated vehicles for safe and efficient movement.Emergency situations may arise due to various reasons, for example an ambulance or a fire truck navigating through traffic needs a “green corridor”, especially when human lives are at stake. With cooperation among the automated vehicles, emergency situations can be handled in an efficient manner. To this end, this diploma thesis investigates the case of an automated Emergency Vehicle (EmV), which aims at driving through traffic, while maintaining its desired speed. Regarding the connectivity capabilities of the EmV, we consider different approaches, where the EmV has no direct interaction with the rest of the traffic in terms of changing their behavior (passive approach); or cases where the EmV is capable of exchanging enhanced information with the surrounding vehicles in order to facilitate its movement, e.g. by creating “green corridors” (active approach).The aforementioned problem is formulated as an Optimal Control Problem (OCP) and is solved numerically with use of a Feasible Direction Algorithm (FDA). The objective function of the OCP aims at minimizing several terms, regarding the safety and comfort of the passengers, fuel efficiency, and advancement goals of EmV by penalizing deviations from longitudinal and lateral desired speeds. State-dependent bounds on control inputs are used to ensure that the vehicles stay within the road boundaries and prevent crashes in emergency situations. The OCP is solved repeatedly for short-time horizons within a model predictive control (MPC) framework, while the vehicle advances.The performance of the EmV is evaluated for both passive and active approaches and for different scenarios and traffic densities. The simulations were conducted in a lane-free environment, using a custom-made extension, named TrafficFluid-Sim, which is built for the SUMO (Simulation of Urban MObility) simulator. The results indicate that, at low-traffic densities, connectivity/cooperation between the EmV and the surrounding traffic has minor effect on the EmV’s performance, as there is sufficient space for manoeuvring. On the other hand, as the traffic density rises, the cooperation seems to be crucial and measures, like the formation of “green corridors”, are demonstrated to be extremely efficient, allowing the EmV to maintain high speeds, despite the limited space.

Available Files

Services