Fig 1 - uploaded by Wonho Suh
Content may be subject to copyright.
In-Vehicle Simulation
Source publication
Current research in applying the Dynamic Data Driven Application Systems (DDDAS) concept to monitor and manage surface transportation
systems in day-to-day and emergency scenarios is described. This work is focused in four, tightly coupled areas. First, a
novel approach to predicting future system states termed ad hoc distributed simulations has be...
Contexts in source publication
Context 1
... a collection of in-vehicle simulations that are interconnected via wireless links and (possibly) wired network infrastructure. Individually, each simulation only models a portion of the traffic network -that which is of immediate interest to the "owner" of the simulator, Figure 1. Collectively, these simulations could be used to create a kind of distributed simulation system with the ability to make forecasts concerning the entire transportation infrastructure as a whole. ...
Context 2
... the distributed simulation is constructed in an "ad hoc" fashion, in much the same way an arbitrary collection of mobile radios join together to form an ad hoc wireless network. The elements of the physical system modeled by different simulators in an ad hoc distributed simulation may overlap, leading to possibly many duplicate models of portions of the system, as seen in Figure 1. Other parts of the system may not be modeled at all. ...
Citations
... Data streams with various update intervals were examined to study which is the best to drive the simulation for an appropriate representation of the actual traffic situation. They also discussed the concept of ad hoc distributed simulations, where individual in-vehicle simulation captures the "near-by" measurement data and models a portion of the traffic network, from which the server constructs an overall picture of the entire network [31]. Some experiment results of the distributed simulation implemented by a cellular automata model and by a VISSIM model are presented in [32]. ...
The benefit of modeling and simulation in rail transit operations has been demonstrated in various studies. However, the complex dynamics involved and the ever-changing environment in which rail systems evolve expose the limits of classical simulation. Changing environmental conditions and second order dynamics challenge the validity of the models and seriously reduce model (re-)usability. This paper discusses the potential benefits and requirements of dynamic data-driven simulation in rail systems. The emphasis is placed on automated model reconfiguration, calibration, and validation through the use of data analysis methods. The rationale and requirements are discussed and a process model for data driven calibration and validation is proposed.
... Retraining a model with new data set may be a direct solution, but it encounters a challenge of high cost of computation and storage, especially in traffic system where the traffic flow data is an endless stream and there are large number of roads to predict. Taken in this sense, traffic system is essentially a Dynamic Data Driven Application Systems (DDDAS)[3] where there is not a fixed model to depict the system accurately. New data are continuously injected into the system and the prediction model should be updated dynamically. ...
Traffic flow prediction is a basic function of Intelligent Transportation System. Due to the complexity of traffic phenomenon, most existing methods build complex models such as neural networks for traffic flow prediction. As a model may lose effect with time lapse, it is important to update the model on line. However, the high computational cost of maintaining a complex model puts great challenge for model updating. The high computation cost lies in two aspects: computation of complex model coefficients and huge amount training data for it. In this paper, we propose to use a nonparametric approach based on locally weighted learning to predict traffic flow. Our approach incrementally incorporates new data to the model and is computationally efficient, which makes it suitable for online model updating and predicting. In addition, we adopt wavelet analysis to extract the periodic characteristic of the traffic data, which is then used for the input of the prediction model instead of the raw traffic flow data. The primary experiments on real data demonstrate the effectiveness and efficiency of our approach.
Artificial intelligence (AI) can contribute to the management of a data driven simulation system, in particular with regard to adaptive selection of data and refinement of the model on which the simulation is based. We consider two different classes of intelligent agent that can control a data driven simulation: (a) an autonomous agent using internal simulation to test and refine a model of its environment and (b) an assistant agent managing a data-driven simulation to help humans understand a complex system (assisted model-building). We present a prototype implementation of an assistant agent to apply DDDAS to social simulations. The automation of the data-driven model development requires content interpretation of both the simulation and the corresponding real-world data. The paper discusses the use of Association Rule Mining to produce general logical statements about simulation and data content as well as the use of logical consistency checking to detect observations that refute the simulation predictions. Finally we consider ways in which this kind of assistant agent can cooperate with autonomous data collection and analysis agents to build a more complete and reliable picture of the observed system.
This is the 5th International Workshop on Dynamic Data Driven Applications Systems (DDDAS), organized in conjunction with ICCS. The DDDAS concept entails the ability to dynamically incorporate data into an executing application simulation, and in reverse, the ability of applications to dynamically steer measurement processes. Such dynamic data inputs can be acquired in real-time on-line or they can be archival data. DDDAS is leading to new capabilities by improving applications modeling and systems management methods, augmenting the analysis and prediction capabilities of simulations, improving the efficiency of simulations and the effectiveness of measurement systems. The scope of the present workshop provides examples of research and technology advances enabled through DDDAS and driven by DDDAS. The papers presented in this workshop represent ongoing multidisciplinary research efforts by an international set of researchers from academe, industry, national and research laboratories.
