Valerio De Martinis

Human-Like Adaptive Cruise Control Systems through a Learning Machine Approach

In this work an Adaptive Cruise Control (ACC) model, with human-like driving capabilities,based on a learning machine approach, is proposed. The system is based on a neural network approach and is intended to assist the drivers in safe car-following conditions. The proposed approach allows for an extreme flexibility of the ACC that can be continuously trained by drivers in order to accommodate their actual driving preferences as these changes among drivers and over time. The model has been calibrated against accurate experimental data consisting in trajectories of vehicle platoons gathered on urban roads. Its performances have been compared with those of a conventional car-following model.

Estimating the effects of energy-efficient driving profiles on railway consumption

In this paper we study the optimisation of railway driving profiles to minimise energy consumption. An optimisation model is proposed for designing the optimal driving profile and a solution procedure. The model and method are tested on some railway sections in order to estimate the potential effects of energy-efficient driving on consumption.

The evolution and planning of hierarchical transport networks

The continuing growth of urban regions coupled with their strong interactions with transportation systems places great emphasis on the coevolution of these systems. In this paper we introduce three developments. First, as a preliminary, the dynamic model of the evolution of retail structure is extended to show how hierarchies emerge. Secondly, these ideas are applied to a model of the evolution in transport networks. Thirdly, it is shown that such a model can be translated into a planning context and the notion of an explicitly hierarchical dynamic model is introduced. An example of the use of such a model of evolution within a network planning context is presented. This explores the planning of network evolution in a case study of Naples arising from a new harbour which is under construction. The model provides a new method for transport system planning through combining the insights from dynamic modelling, particularly with respect to the sequencing of investment, with the practicalities of budget constraints.

Improving energy efficiency for freight trains during operation: The use of simulation

This paper focuses on the optimization of freight trains operation for energy saving purposes, highlighting the future challenges and the current possibilities. A simulation-based approach for the generation of energy saving speed profiles, implemented in Matlab, is here adopted with real data from onboard monitoring system. Results have highlighted two interesting aspects: onboard monitoring data give useful information during the calibration of the simulation model, by referring to real data it is possible to have a clear reference on the potential energy saving.

Enhancing energy efficiency in railway operation through RCG-based rescheduling

This paper presents a new approach for train rescheduling, based on a Resource Conflict Graph (RCG) approach, which considers energy efficiency aspects into solutions. The model is expressly developed for highly congested mixed traffic networks, where rail operation deals with different rail services (Intercity, InterRegio; single wagonload, mail trains) and high computing speed is required. Here energy consumption is considered as a constraint, in order to improve algorithm performances and, at the same time, to obtain a more energy efficient solution. First experiments with real data of a Swiss network show how the model behaves with different energy consumption constraints and its suitability for real time applications.

Energy savings in mixed rail traffic rescheduling: an RCG approach

This paper presents an enhanced version of the Resource Conflict Graph (RCG) approach for railway rescheduling that considers the reduction of tractive energy consumption in the rescheduling process. The objectives and constraints considered in freight train rescheduling are different from those related to passenger trains. For example, there is generally more flexibility in freight train scheduling regarding routes and departure/arrival times. On the other hand, freight trains have frequent unplanned stops for overtaking and acceleration phases use significant energy, especially for long freight trains. Therefore, energy consumption is recognized as an important cost factor in freight train operations and it is commonly used to help plan schedules. Energy consumption can be reduced by choosing energy efficient paths (less traffic, less gradient) and driving with speed profiles that improve route regularity (i.e., that avoid unplanned stops and minimize acceleration phases). The paper describes the proposed rescheduling approach and then applies it in a numerical example evaluated using the test track of the Railway Operations Lab located in the facilities of the Institute for Transport Planning and Systems (IVT) at ETH Zurich.