The nature of the social relationship within a pedestrian group influences the group’s structure and behavior and thus its entitativity (i.e. the perception of a group as a unit by other pedestrians). However, existing crowd models ignore the diversity of social relationships and have limitations in reproducing group avoidance behaviors. The proposed model is an adaptation of the social force model that addresses group social relationships. The approach is calibrated by comparing the distances and angles between members of the simulated groups with observations in real crowds. Results show that intra-group distances are a key factor in collision avoidance behavior. Simulation of collision avoidance shows that group members behavior fits better with empirical data than the original model and that individuals avoid splitting groups. By simply tuning the distribution of social relationships in the simulated crowd, the model can be used to reproduce crowd behaviors in several contexts.
Demand response (DR) has gained a significant recent interest due to its potential for mitigating many power system problems. Game theory is a very effective tool to be utilized in DR management. In this paper, the DR between a distribution system operator (DSO) and load aggregators (LAs) is designed as a Stackelberg game, where the DSO acts as the leader and LAs are regarded as the followers. Due to the limitations of the centralized solution approaches, a genetic algorithm-based decentralized approach is proposed. To demonstrate the proposed approach, a case study concerning a day-ahead optimization for a real-time pricing market with a single DSO and three LAs is designed and optimized. The proposed approach is able to shift the demand peaks and prove that it has a great potential to be used for the Stackelberg game between a DSO and multiple LAs to fully exploit the potential of DR.
Large organizations now use Modeling and Simulation (M&S) for complex systems. This evolution has also resulted in a lack of simple tools and processes to define models and experiments which hinders wide scale user adoption of M&S to solve many real-world problems. By exploring the Arctic logistical support base problem, we will illustrate how a simple M&S process and system can be used by operators with limited knowledge of M&S. This will be demonstrated by determining a temporary logistical support base for re-supplying military operations in the Canadian Arctic.
National security decisions are driven by complex, interconnected contextual, individual, and strategic variables. Modeling and simulation tools are often used to identify relevant patterns, which can then be shaped through policy remedies. In the paper to follow, however, we argue that models of these scenarios may be prone to the complexity-scarcity gap, in which relevant scenarios are too complex to model from first principles and data from historical scenarios are too sparse---making it difficult to draw representative conclusions. The result are models that are either too simple or are unduly biased by the assumptions of the analyst. We outline a new method of quantitative inquiry---experimental wargaming---as a means to bridge the complexity-scarcity gap that offers human-generated, empirical data to inform a variety of model and simulation tasks (model building, calibration, testing, and validation). Below, we briefly describe SIGNAL---our first-of-a-kind experimental wargame designed to study strategic stability in conflict settings with nuclear weapons. We then highlight the potential utility of this data for modeling and simulation efforts in the future using this data.
The progressive integration of renewable energy resources in the modern power grid can result in unsatisfactory frequency responses. To address this problem, wind turbine generators (WTGs), in particular, can be employed to support the power grid. Most works, however, only consider a simple step disturbance. In this paper, a new class of worst-case disturbances is introduced and their nefarious impact on renewable-penetrated power systems is investigated. The worst-case disturbances are derived using optimal control theory. More importantly, the proposed mechanism allows to determine the reaction time to trigger supportive control actions of WTGs to ensure satisfactory frequency response. Numerical results are provided and the effects of the worst-case disturbance on the reaction time are presented.
Multiobjective optimization problems (MOPs) are common across many science and engineering fields. A multiobjective optimization algorithm (MOA) seeks to provide an approximation to the tradeoff surface between multiple, possibly conflicting, objectives. Many MOPs are the result of objective functions that require the evaluation of a computationally expensive numerical simulation. Solving these large and complex problems requires efficient coordination between the MOA and the computationally expensive cost functions. In this work, a recently proposed MOA is integrated into the libEnsemble software library, which coordinates extreme scale resources for large ensemble computations. Efficient integration requires fundamental changes to the underlying MOA. The convergence and performance results for the integrated and original MOA are compared on a set of benchmark problems.
Scoliotic deformities may be addressed with either anterior or posterior approaches for scoliosis correction procedures. While typically quite invasive, the impact of these operations may be reduced through the use of computer-assisted surgery. A combination of physician-designated anatomical landmarks and surgical ontologies allows for real-time intraoperative guidance during computer-assisted surgical interventions. Predetermined landmarks are labeled on an identical patient model, which seeks to encompass vertebrae, intervertebral disks, ligaments, and other soft tissues. The inclusion of this anatomy permits the consideration of hypothetical forces that are previously not well characterized in a patient-specific manner. Updated ontologies then suggest procedural directions throughout the surgical corridor, observing the positioning of both the physician and the anatomical landmarks of interest at the present moment. Merging patient-specific models, physician-designated landmarks, and ontologies to produce real-time recommendations magnifies the successful outcome of scoliosis correction through enhanced pre-surgical planning, reduced invasiveness, and shorted recovery time.
Autonomous mobile robots depend heavily on sensors for interpreting the external environment and planning their movement. Given the complexity of even simple environments, the use of simulation is necessary to test sensory perception and movement strategies efficiently. Accurate and efficient sensor simulation is key to the successful development and fielding of autonomous mobile robots. In this paper, we investigate the scalability of LIDAR-type sensor simulation in ROS-supported Gazebo, an open-source multi-robot simulator. Specifically, the performance and scalability of the simulator are assessed with and without support from GPU in two virtual worlds, one highly-complex in terms of the number of collision targets and one not-complex. In both cases, we increase the number of sensors and evaluate system performance. The results show that the use of GPU helps improve performance, providing better speedups in the complex scene. However, the number of sensors that can be used is limited when using GPU.
In this paper, we revisit the method of a phase transition model for representing hybrid systems and temporal logic specifications (TLSs) for specifying desired behaviors of systems, and discuss their usefulness for smart energy systems. On the one hand, the phase transition model incorporates the continuous model of the relay device action with a particular structural form that allows for the construction of a single, global differential-algebraic equation for hybrid systems (thus smoothed hybrid systems). On the other hand, the TLS allows sophisticated descriptions of control specifications addressing both magnitude and time simultaneously, which has recently been applied to the control strategy of several types of continuous and hybrid systems. We provide high-level descriptions of each of the two techniques and present simulation results in the context of smart energy systems.
Distributed Ledger Technology (DLT) utilizes an architecture that can host a large number of nodes without pre-established trust to provide decentralized services. The blockchain is the most widely used architecture of distributed ledger, where transactions across the whole network are visible to all participants in a chain to prevent tampering. However, transactions may contain sensitive information such as business contract. To secure the system and protect user privacy, we propose a multi-channel architecture that leverages Intel Software Guard Extensions (SGX). We illustrate how SGX capabilities help to defend against attacks on distributed ledgers, by way of SGX enforcement on the participating machines. We adopt the design and implementation of a two layer architecture for securing the blockchain mining process and enhancing the transaction privacy. The security analysis and performance evaluation show that the design and protocols are capable of protecting privacy, defending against adversarial attacks and scalable.
The vast majority of recommender system research has focused on improving performance accuracy, while limited work has explored their societal, network level effects. This paper demonstrates how simulation can be used to investigate macro level effects of online social network link recommendations, such as whether these technologies may be fragmenting or bridging communities of individuals. An agent-based model is presented that generates stylized online social networks with different percentages of real world contacts and link recommendations. Results show that networks with higher percentages of recommendation-based links produce more clustered, distinct, and dispersed communities, suggesting that these technologies could fragment society. Furthermore, scale-free network properties diminished with higher percentages of recommendations, suggesting that these technologies could be contributing to recent findings that social networks are at most `weakly' scale-free. Building upon this research, further simulation work could inform the design of link recommendation algorithms that help connect both individuals and communities.
Recent disasters have shown that hazards can be unpredictable and can have catastrophic consequences. Emergency plans are key to dealing with these situations and communications play a key role in emergency management. In this paper, we provide a formalism to design resilient emergency plans in terms of communications. We exemplify how to use the formalism using a case study of a Nuclear Emergency Plan.
In this paper, we outline the construction and initial simulation experiment results of the Marginalization model (MARG). We experiment under different group parameters because the theoretical paradigm we follow views bullying as a result of social processes. Our primary research question explores the possibility of bullying emergence as agents select interaction partners in a university setting. Based on the simulated process, our results take indications of the stress of marginalization in a student group as a proxy for emergent marginalization. MARG simulates two types of interactions between pairs of students: forced and hang-out interactions. In the latter, students decide whether to interact based on individual preferences formed by social norms and individual tolerance related to those norms. The emergence of intensified marginalization from MARG processes leads to some ethical considerations and provides ground for discussions concerning suitable interventions.
We present auto_diff, a package that performs automatic differentiation of numerical Python code. auto_diff overrides Python's NumPy package's functions, augmenting them with seamless automatic differentiation capabilities. Notably, auto_diff is non-intrusive, i.e., the code to be differentiated does not require auto_diff-specific alterations. We illustrate auto_diff on electronic devices, a circuit simulation, and a mechanical system simulation. In our evaluations so far, we found that running simulations with auto_diff takes less than 4 times as long as simulations with hand-written differentiation code. We believe that auto_diff, which was written after attempts to use existing automatic differentiation packages on our applications ran into difficulties, caters to an important need within the numerical Python community. We have attempted to write this paper in a tutorial style to make it accessible to those without prior background in automatic differentiation techniques and packages. We have released auto_diff as open source on GitHub.
We present an approach to studying human behavior in the Iterated Prisoner's Dilemma (IPD) game, where human participants play against software bots in an online environment. This setting allows precise control of the strategies faced by the human subjects. Our goal is to test models of human behavior and the means by which cooperation can be promoted. Results from our experiments with subjects from Amazon Mechanical Turk show that the leading models of human behavior in the IPD are incomplete, as they can explain our data only partially. We propose a new model of behavior, Majority Wins, which provides a better fit to the data, which we test and demonstrate through a simulation.
Mobile Adhoc Network-based (MANET-based) infrastructures have been proposed as communication support in case of disastrous events and emergency situations. Indeed, standard communication systems like cellular networks might become unavailable or entirely destroyed. However, designing a reliable network which various heterogeneous systems will operate on is a well-known and costly problem. Simulation represents a good alternative for checking properties on these networks, but simulators usually lack confidence or accuracy. Therefore, an integrated simulator architecture of MANET-based network has been proposed. This paper illustrates the implementation of this kind of architecture, and discuss the pros and cons of this kind of solution.
The North American bulk power system is one of the most vital infrastructures in modern society as it accounts for virtually all the electricity supplied to the United States, Canada, and a portion of Baja Notre California, Mexico. Cyberattacks, of all forms, are becoming increasingly prominent within power networks and other infrastructures whereas their resolution can consume a significant deal of time and monetary resources. This can be further worsened if there are subsequent physical attacks in the wake of a cyberattack, as the system downtime leaves the government, military, and other critical infrastructures incredibly vulnerable This research aims to investigate if different patterns of cyberattacks could be identified with speed using simulation and machine learning algorithms. More specifically, we design a simulation model that can help better defend against cyber threats.
Connected autonomous vehicles are an important class of cyber-physical systems that are expected to have a major impact on society. Connectivity and autonomy in next-generation automobiles can be leveraged to improve safety and efficiency of our transportation systems. This paper presents a novel approach for incorporating individual driving preferences of vehicles in a computational framework to allow dynamic assignment and transfer of right-of-way privileges between cars as they navigate contested road segments. Dynamic priorities based on time of arrival estimates and positions in queues are used to unambiguously identify the owners of right-of-way privileges to conflict zones at any given time. A mechanism for transferring the privileges from a unique rightful owner to another car, possibly incentivized by using a shared currency, is proposed. A simulation framework using MATLAB® is developed to enable rigorous study of this mechanism across tens of thousands of simulations.
High frequency oscillations (HFOs) have been used for seizure prediction and are promising biomarkers of epileptogenesis. However, detecting HFOs is time consuming and subjective, prompting research into automated detection and classification pipelines. We aim to understand how different EEG filtering methods impact these pipelines and harmonize detections from the same data when preprocessed differently. We preprocessed EEG with two different filters and then detected events with the short time energy (STE) detector and compared common detections. We applied t-distributed stochastic neighbor embedding (t-SNE) to the datasets and compared embeddings then investigated if shifting commonly detected events prior to t-SNE helped standardize embeddings. The finite impulse response (FIR) and infinite impulse response (IIR) filters achieved a Cohen's Kappa coefficient of 0.8962 after shifting, reflecting a high level of agreement. The t-SNE embeddings were similar only when data were shifted prior to embedding. Feasible solutions to this shifting problem are addressed.
The Internet of Things (IoT) devices such as sensors and video cameras have small memories and less computational power. The video analytics of traditional approaches for detection, tracking and pattern recognition of moving objects used only in the cloud. This approach suffered from high latency and more network bandwidth to transfer data into the cloud. We address this problem by using edge computing devices between IoT devices and the cloud. We propose a new framework for scalable object detection, tracking and pattern recognition of moving objects that relies on dimensionality reduction with edge computing architecture. We also propose a scalable object detection and tracking method based on You Only Look Once (YOLO) method. The experiment demonstrates that our proposed method will save network bandwidth and processing time. The performance of object detection and tracking model is greater than 96%. This shows that our method has greater performance than existing models.
