Institutional Repository [SANDBOX]
Technical University of Crete
EN  |  EL

Search

Browse

My Space

Effect of automated vehicles on highway traffic flow and signalized junctions

Typaldos Panagiotis

Full record


URI: http://purl.tuc.gr/dl/dias/7217591D-1AB2-4AC2-BEB1-285907F15983
Year 2022
Type of Item Doctoral Dissertation
License
Details
Bibliographic Citation Panagiotis Typaldos, "Effect of automated vehicles on highway traffic flow and signalized junctions", Doctoral Dissertation, School of Production Engineering and Management, Technical University of Crete, Chania, Greece, 2022 https://doi.org/10.26233/heallink.tuc.94451
Appears in Collections

Summary

Over the past years, intensive research has been carried out towards the development of fully automated driving for road vehicles. Fully automated vehicles could improve safety and efficiency of traffic operations by reducing accidents caused by human driver errors; improving driver and passenger comfort; reducing traffic congestion and, at the same time, they contribute to the environmental and economical aspects by reducing fuel consumption and emissions. Although vehicle automation has already led to significant achievements in supporting the driver in various ways, rising the level of automation to fully-automated driving is an extremely challenging problem. This is mainly due to the complexity of real-world environments, including avoidance of static and moving obstacles, compliance with traffic rules and consideration of human driving behavior aspects. The importance of eco-driving in reducing cumulative fuel consumption of road vehicles is a well-known and widely treated subject. Eco-driving intends to minimize fuel consumption by maneuvering a vehicle with a human or automated driver. In the current work, the eco-driving problem is cast in an optimal control framework with fixed time horizon. For the fuel consumption estimation, a number of alternatives are employed. Initially, a realistic, but nonlinear and non-smooth formula from the literature is considered. Simple smoothing procedures are then applied, so as to enable the application of powerful numerical algorithms for the efficient solution of the resulting nonlinear optimal control problem. Furthermore, suitable quadratic approximations of the nonlinear formula are also considered, which enable analytical problem solutions. A comprehensive comparison on the basis of various driving scenarios demonstrates the features of each alternative approach. In particular, it is demonstrated that the often utilized, but sometimes strongly questioned, square-of-acceleration term delivers excellent approximations for fuel minimizing trajectories in the present setting. Despite the, seemingly, simple structure of highway environments, human drivers' actions may, often, lead to accidents and excessive fuel consumption. Automated vehicles equipped by appropriate sensor and path-planning algorithms have the potential to overcome these issues. In this context, a path-planning algorithm for connected and non-connected automated road vehicles on multilane motorways is derived from the opportune formulation of an optimal control problem. In this framework, the objective function to be minimized contains appropriate respective terms to reflect: the goals of vehicle advancement; passenger comfort; and avoidance of collisions with other vehicles and of road departures. Connectivity implies, within the present work, that connected vehicles can exchange with each other (V2V) real-time information about their last generated short-term path. For the numerical solution of the optimal control problem, an efficient feasible direction algorithm (FDA) is used. To ensure high-quality local minima, a simplified Dynamic Programming (DP) algorithm is also conceived to deliver the initial guess trajectory for the start of the FDA iterations. Thanks to very low computation times, the approach is readily executable within a Model Predictive Control (MPC) framework. The proposed MPC-based approach is embedded within a microsimulation platform, which enables the evaluation of a plethora of realistic vehicle driving and advancement scenarios under different vehicles mixes. Results obtained on a multilane motorway stretch indicate higher efficiency of the optimally controlled vehicles in driving closer to their desired speed, compared to ordinary manually driven vehicles. Increased penetration rates of automated vehicles are found to increase the efficiency of the overall traffic flow, benefiting manual vehicles as well. Moreover, connected controlled vehicles appear to be more efficient in achieving their desired speed, compared also to the corresponding non-connected controlled vehicles, due to the improved real-time information and short-term prediction achieved via V2V communication. Urban areas, and specifically signalized intersections have a crucial role on the safety of drivers, as well as the increased fuel consumption of vehicles. To address these issues, appropriate systems have been developed, known as Green Light Optimal Speed Advisory (GLOSA) systems. In the current work, a discrete-time stochastic optimal control problem has been proposed to address the GLOSA problem in cases where the next signal switching time is decided in real time and is therefore uncertain in advance. The corresponding numerical solution via SDP (Stochastic Dynamic Programming) calls for substantial computational time, which excludes problem solution in the vehicle's on-board computer in real time. To overcome the computation time bottleneck, several modified versions of Dynamic Programming were also developed. As a first attempt, a Discrete Differential Dynamic Programming (DDDP) was employed for the numerical solution of the stochastic optimal control problem. The DDDP algorithm was demonstrated to achieve results equivalent to those obtained with the ordinary SDP algorithm, albeit with significantly reduced computational times. Subsequently, a different modified version of Dynamic Programming, known as Differential Dynamic Programming (DDP) was utilized. For the stochastic GLOSA problem, it is demonstrated that DDP achieves quasi-instantaneous (extremely fast) solutions in terms of CPU times, which allows for the proposed approach to be readily executable online, within an MPC framework, in the vehicle's on-board computer. The approach is demonstrated by use of realistic examples. It should be noted that DDP does not require discretization of variables, hence the obtained solutions may be slightly superior than the standard SDP solutions. However, in all pre-mentioned GLOSA problems, there was an assumption that the traffic signal is initially red and turns to green, which means that only half traffic light cycle was considered. To this end, the aforementioned problem was extended considering a full traffic light cycle, consisting of four phases: a certain green phase, during which the vehicle can freely pass; an uncertain green phase, in which there is a probability that the traffic light will extend its duration or turn to red at any time; a certain red phase during which the vehicle cannot pass; and an uncertain red phase, in which there is a probability that the red signal may be extended or turn to green at any time. Preliminary results indicate that the proposed SDP approach achieves better average performance, in terms of fuel consumption, compared to the IDM (Intelligent Driver Model) model, which emulates human-driving behavior.

Available Files

Services

Statistics