Search

The modeling and simulation community can benefit from building and running (co-)simulation models on-demand. In order to drive innovation further, it should be easy to set up, orchestrate, and execute simulations – even for researchers with no background in computer science. Additionally, open interfaces should be the default to enable a variety of applications. The possibility to connect external components smoothly with a simulation run allows for elaborate experiments. In this work, the details of the architecture and design of a simulation platform enabling evaluations of future mobility scenarios are presented. While the explanations are based on mobility applications, the concept can be generalized and applied to other domains. The novelty lies in the use of a big data stream processing platform called Apache Kafka. All kinds of required communication are handled by the Kafka middleware. Even the coupling of different simulators is realized on top of it. As a foundation, existing works regarding Modeling and Simulation as a Service are presented. The approach is then described by its architecture, its interfaces, and the life-cycle of a simulation run. Further information is given on the distribution of simulations and the topic-based clock synchronization mechanism. A performance evaluation shows that the approach is on one level with state-of-the art co-simulation middlewares.

In this paper, we evaluate the n-ray ground interference model based on field experiments. The n-ray model is an extension of the popular two-ray interference model for three-dimensional Vehicular Ad Hoc Network (VANET) simulation. Using commodity hardware, we measure the path loss on nine selected test tracks and replay the scenarios simulatively to compare it to the simulated path loss resulting from the n-ray model. The simulation results are in good agreement with the measurements for the most parts in all nine cases, indicating that the n-ray ground interference model is able to capture the effect of ground reflections for arbitrary terrain shapes. The two-ray interference model fails to do so as it neglects elevation information. Moreover, the dependence of the results on the parameterization of the model as well as on the horizontal resolution of the underlying elevation dataset is investigated. In order to ensure that the n-ray model is only applied under Line of Sight (LOS) conditions, we further present a possible combination with our environmental diffraction model.

The High Level Architecture is an IEEE standard that enables distributed simulation. There are several implementations of the standard, which are providing i.a. the Run-Time Infrastructure component. This paper compares the four most known RTI Implementations, namely MAK RTI, Pitch pRTI, Portico, and CERTI, with a focus on performance evaluation. In general, Pitch pRTI was the fastest implementation for most of our experiments. CERTI performed best for big payload sizes and Portico showed an interesting oscillation pattern.

Typically, there is more than one possible answer to the question of how to model and simulate a problem. Combinations of different simulation paradigms and techniques are currently in vogue and potentially necessary to meet certain requirements. Besides clear advantages, model combinations tend to increase the complexity of the overall model building process. Even if different paradigms are coupled via a middleware, another use case might require reimplementing the connections. Additionally, users need increased effort to understand how the results have been generated in composed models. If it is not clear how data has been transformed from input to output, the trust in simulation is impaired. In order to counteract, a concept for reusable translation between submodels is proposed, leading to a better transparency regarding the data flows. We applied the High Level Architecture based approach to connect a macroscopic, a mesoscopic, and a microscopic traffic simulation.

The increasing precision in simulation models leads to a high computational effort. In general, parallel and distributed simulation is one technique to speed up a simulation to an acceptable runtime. Depending on the simulation model, the suitability of a parallel or distributed simulation varies. This applies also to coupled V2X simulations, where distributed computation may help to increase performance drastically. In this work, we identify opportunities and possibilities regarding distributed V2X simulations. Therefore, we describe and evaluate an approach to distribute multiple instances of the well-known Veins framework that are interacting and running concurrently. The instances are coupled using a hybrid co-simulation framework based on the High Level Architecture. With increasing traffic load, a higher speedup can be measured.

Due to the increase of the complexity and uncertainty in the future sustainable energy system new control algorithms for decentralized acting energy entities are needed. We present an approach of distributed Reinforcement Learning in a multi-agent setup to find a control strategy of two cooperative agents within an energy cell. In order to practice energy sharing to decrease the energy cell's overall interdependence on the electrical grid, we train two independently learning agents, an energy storage and an electric power generator using Q-learning. We compare the learned strategy of the agents under partial and full observability of the environment and evaluate the interdependence of the energy cell on the electrical grid. Our results show that distributed Q-learning with independently learning agents works in the setup of an energy cell without the necessity of information exchange between agents. The algorithm under partial observability of the environment reaches comparable performance to that of full observability with fewer need of communication but at the cost of five times longer training time.

This paper demonstrates that interferometry is an effective means to provide elevation information in an automotive radar. Often discarded as yielding ‘no real resolution in elevation’, we display experimental results showing that high resolution in range, velocity and azimuth renders resolution in elevation dimension redundant.

Millimeter-wave (mmW) sensorics are an interesting option for several short range applications like the imaging of persons or the monitoring of indoor environments. Due to the small antenna dimensions, compact sensor setups can be realised. This paper will discuss different setups for mmW-sensors based on the principle of a frequency-modulated continuous wave radar with respect to system architecture, calibration concepts, resulting sensor capabilities and imaging results.