Crowd Simulation Algorithms Research Articles

Differentiable Learning of Scalable Multi-Agent Navigation Policies

We present an end-to-end differentiable learning algorithm for multi-agent navigation policies. Compared with prior model-free learning algorithms, our method leads to a significant speedup via the gradient information. Our key innovation lies in a novel differentiability analysis of the optimization-based crowd simulation algorithm via the implicit function theorem. Inspired by continuum multi-agent modeling techniques, we further propose a kernel-based policy parameterization, allowing our learned policy to scale up to an arbitrary number of agents without re-training. We evaluate our algorithm on two tasks in obstacle-rich environments, partially labeled navigation and evacuation, for which loss functions can be defined making the entire task learnable in an end-to-end manner. The results show that our method can achieve more than one order of magnitude speedup over model-free baselines and readily scale to unseen target configurations and agent sizes.

Dynamic leader role modeling for self‐organizing crowd evacuation simulation

AbstractThe role of the leader in the self‐organizing crowd has a significant influence on the crowd evacuation when an emergency occurs. However, current crowd evacuation models either ignore the leadership characteristics of pedestrians or predefine an invariable leader identity. This article clearly distinguishes leaders from followers in a crowd and evaluates the impact of the leader's role on crowd evacuation under dynamic situations. First, the leadership model of the pedestrian is built based on the classic OCEAN personality parameters. Social dominance tendency is measured for more credibility. Then, the dynamic leader role model (DLRM) is proposed to describe the relationship between leaders and followers, considering the interaction among pedestrians during the evacuation. Finally, a novel crowd simulation algorithm based on the above DLRM is presented using the extended social force model. Various experiments in several typical scenarios verified that our proposed simulation model has better realism.

The Design and Optimization of Ship Cabin Space Layout Based on Crowd Simulation

<p indent=0mm>In order to solve the space layout optimization problem of ship cabin efficiently, an interactive layout design and optimization method based on crowd simulation and tensor field is proposed. Firstly, the image based flow visualization algorithm is used to analyze the tensor field of the space layout domain, which calculates the eigenvector and generates the road network of ship, and obtain the cabin space layout based on the road network. Secondly, the crowd simulation algorithm is used to simulate the behavioral activities of the crowd in the cabin space, and then to evaluate the mobility and coziness of the crowd in the ship cabin space. After obtaining a series of road network of ship and results of crowd simulation, the mapping relationship between road width and road spacing in road network and mobility value and comfort value in crowd simulation results is established respectively. The results of crowd simulation are used as evaluation criterion to interactively and iteratively adjust the characteristic parameters of tensor field. The experimental results show that the proposed method effectively establishes the mapping relationship between tensor field and crowd simulation. Compared with genetic algorithm and analytic hierarchy process, it can quickly and efficiently design high-quality cabin space layout in line with human activities demand.

IMPLEMENTATION OF A GENERIC FRAMEWORK ON CROWD SIMULATION: A NEW ENVIRONMENT TO MODEL CROWD BEHAVIOR AND DESIGN VIDEO GAMES

Crowd behavior is the collective act and gathering of a group of individuals to achieve a shared purpose. Swarm intelligence-based optimization algorithms are usually used to solve complex problems for crowd behavior. Crowd simulations are often used for the analyses that require precision in different domains such as complex structural analysis, image recognition, creating nature-inspired non-player character movements in video games, and more. In this study, a generic crowd simulation framework that can be used to simulate already-available crowd simulation algorithms and design new ones was developed. The test environment layout was generated with the use of a generate-and-test algorithm combined with the crowd simulation algorithms to make sure that the generated content is meeting the requirements of a crowd simulation environment. Within the framework, three different crowd simulation algorithms —firefly algorithm, particle swarm optimization, and artificial bee colony— are generated and also implemented as puzzle-like video games. The results show that all fireflies achieved to gather at the global minimum of the generated layout faster and in a more precise way than the artificial bee colony algorithm and particle swarm optimization algorithm. The developed framework enables a generic and parametric testbed to design and compare different algorithms and to generate video games.

Crowd Navigation in VR: Exploring Haptic Rendering of Collisions.

Virtual reality (VR) is a valuable experimental tool for studying human movement, including the analysis of interactions during locomotion tasks for developing crowd simulation algorithms. However, these studies are generally limited to distant interactions in crowds, due to the difficulty of rendering realistic sensations of collisions in VR. In this article, we explore the use of wearable haptics to render contacts during virtual crowd navigation. We focus on the behavioral changes occurring with or without haptic rendering during a navigation task in a dense crowd, as well as on potential after-effects introduced by the use haptic rendering. Our objective is to provide recommendations for designing VR setup to study crowd navigation behavior. To the end, we designed an experiment (N=23) where participants navigated in a crowded virtual train station without, then with, and then again without haptic feedback of their collisions with virtual characters. Results show that providing haptic feedback improved the overall realism of the interaction, as participants more actively avoided collisions. We also noticed a significant after-effect in the users' behavior when haptic rendering was once again disabled in the third part of the experiment. Nonetheless, haptic feedback did not have any significant impact on the users' sense of presence and embodiment.

A Hybrid Algorithm in Reinforcement Learning for Crowd Simulation

Exploiting the efficiency and stability of Dynamic Crowd, the paper proposes a hybrid crowd simulation algorithm that runs using multi agents and it mainly focuses on identifying the crowd to simulate. An efficient measurement for both static and dynamic crowd simulation is applied in tracking and transportation applications. The proposed Hybrid Agent Reinforcement Learning (HARL) algorithm combines the Q-Learning off-policy value function and SARSA algorithm on-policy value function, which is used for dynamic crowd evacuation scenario. The HARL algorithm performs multiple value functions and combines the policy value function derived from the multi agent to improve the performance. In addition, the efficiency of the HARL algorithm is able to demonstrate in varied crowd sizes. Two kinds of applications are used in Reinforcement Learning such as tracking applications and transportation monitoring applications for pretending the crowd sizes.

Coordinating Multi-Agent Navigation by Learning Communication

This work presents a decentralized multi-agent navigation approach that allows agents to coordinate their motion through local communication. Our approach allows agents to develop their own emergent language of communication through an optimization process that simultaneously determines what agents say in response to their spatial observations and how agents interpret communication from others to update their motion. We apply our communication approach together with the TTC-Forces crowd simulation algorithm (a recent, high performing, anticipatory collision technique) and show a significant decrease in congestion and bottle-necking of agents, especially in scenarios where agents benefit from close coordination. In addition to reaching their goals faster, agents using our approach show coordinated behaviors including greeting, flocking, following, and grouping. Furthermore, we observe that communication strategies optimized for one scenario often continue to provide time-efficient, coordinated motion between agents when applied to different scenarios. This suggests that the agents are learning to generalize strategies for coordination through their communication "language".

A human and group behavior simulation evaluation framework utilizing composition and video analysis

AbstractIn this work, we present the modular crowd simulation evaluation through composition framework, which provides a quantitative comparison between different pedestrian and crowd simulation approaches. Evaluation is made based on the comparison of source footage against synthetic video created through novel composition techniques. The proposed framework seeks to reduce the complexity of simulation evaluation and provide a platform from which the comparison of differing simulation algorithms and parametric tuning can be conducted to improve simulation accuracy or provide measures of similarity between crowd simulation algorithms and source data. Through the use of features designed to mimic the human visual system, specific simulation properties can be evaluated relative to sample footage. Validation was performed on a number of popular crowd data sets and through comparisons of multiple pedestrian and crowd simulation algorithms.

Gradient-based steering for vision-based crowd simulation algorithms

Most recent crowd simulation algorithms equip agents with a synthetic vision component for steering. They offer promising perspectives through a more realistic simulation of the way humans navigate according to their perception of the surrounding environment. In this paper, we propose a new perception/motion loop to steering agents along collision free trajectories that significantly improves the quality of vision-based crowd simulators. In contrast with solutions where agents avoid collisions in a purely reactive binary way, we suggest exploring the full range of possible adaptations and retaining the locally optimal one. To this end, we introduce a cost function, based on perceptual variables, which estimates an agent's situation considering both the risks of future collision and a desired destination. We then compute the partial derivatives of that function with respect to all possible motion adaptations. The agent then adapts its motion by following the gradient. This paper has thus two main contributions: the definition of a general purpose control scheme for steering synthetic vision-based agents; and the proposition of cost functions for evaluating the perceived danger of the current situation. We demonstrate improvements in several cases.

Detection of Global and Local Motion Changes in Human Crowds

Crowds arise in a variety of situations, such as public concerts and sporting matches. In typical conditions, the crowd moves in an orderly manner, but panic situations may lead to catastrophic results. We propose a computer vision method to identify motion pattern changes in human crowds that can be related to an unusual event. The proposed approach can identify global changes, by evaluating 2D motion histograms in time, and also local effects, by identifying clusters that present similar spatial locations and velocity vectors. The method is tested both on publicly available data sets involving crowded scenarios and on synthetic data produced by a crowd simulation algorithm, which allows the creation of controlled environments with known motion patterns that are particularly suitable for multicamera scenarios.

Going Through, Going Around: A Study on Individual Avoidance of Groups.

When avoiding a group, a walker has two possibilities: either he goes through it or around it. Going through very dense groups or around huge ones would not seem natural and could break any sense of presence in a virtual environment. This paper aims to enable crowd simulators to handle such situations correctly. To this end, we need to understand how real humans decide to go through or around groups. As a first hypothesis, we apply the Principle of Minimum Energy (PME) on different group sizes and density. According to this principle, a walker should go around small and dense groups whereas he should go through large and sparse groups. Such principle has already been used for crowd simulation; the novelty here is to apply it to decide on a global avoidance strategy instead of local adaptations only. Our study quantifies decision thresholds. However, PME leaves some inconclusive situations for which the two solutions paths have similar energetic costs. In a second part, we propose an experiment to corroborate PME decisions thresholds with real observations. As controlling the factors of an experiment with many people is extremely hard, we propose to use Virtual Reality as a new method to observe human behavior. This work represents the first crowd simulation algorithm component directly designed from a VR-based study. We also consider the role of secondary factors in inconclusive situations. We show the influence of the group appearance and direction of relative motion in the decision process. Finally, we draw some guidelines to integrate our conclusions to existing crowd simulators and show an example of such integration. We evaluate the achieved improvements.

Generating pedestrian trajectories consistent with the fundamental diagram based on physiological and psychological factors.

Pedestrian crowds often have been modeled as many-particle system including microscopic multi-agent simulators. One of the key challenges is to unearth governing principles that can model pedestrian movement, and use them to reproduce paths and behaviors that are frequently observed in human crowds. To that effect, we present a novel crowd simulation algorithm that generates pedestrian trajectories that exhibit the speed-density relationships expressed by the Fundamental Diagram. Our approach is based on biomechanical principles and psychological factors. The overall formulation results in better utilization of free space by the pedestrians and can be easily combined with well-known multi-agent simulation techniques with little computational overhead. We are able to generate human-like dense crowd behaviors in large indoor and outdoor environments and validate the results with captured real-world crowd trajectories.

A Data‐Driven Framework for Visual Crowd Analysis

AbstractWe present a novel approach for analyzing the quality of multi‐agent crowd simulation algorithms. Our approach is data‐driven, taking as input a set of user‐defined metrics and reference training data, either synthetic or from video footage of real crowds. Given a simulation, we formulate the crowd analysis problem as an anomaly detection problem and exploit state‐of‐the‐art outlier detection algorithms to address it. To that end, we introduce a new framework for the visual analysis of crowd simulations. Our framework allows us to capture potentially erroneous behaviors on a per‐agent basis either by automatically detecting outliers based on individual evaluation metrics or by accounting for multiple evaluation criteria in a principled fashion using Principle Component Analysis and the notion of Pareto Optimality. We discuss optimizations necessary to allow real‐time performance on large datasets and demonstrate the applicability of our framework through the analysis of simulations created by several widely‐used methods, including a simulation from a commercial game.

Parameter estimation and comparative evaluation of crowd simulations

AbstractWe present a novel framework to evaluate multi‐agent crowd simulation algorithms based on real‐world observations of crowd movements. A key aspect of our approach is to enable fair comparisons by automatically estimating the parameters that enable the simulation algorithms to best fit the given data. We formulate parameter estimation as an optimization problem, and propose a general framework to solve the combinatorial optimization problem for all parameterized crowd simulation algorithms. Our framework supports a variety of metrics to compare reference data and simulation outputs. The reference data may correspond to recorded trajectories, macroscopic parameters, or artist‐driven sketches. We demonstrate the benefits of our framework for example‐based simulation, modeling of cultural variations, artist‐driven crowd animation, and relative comparison of some widely‐used multi‐agent simulation algorithms.

A Whole Surface Approach to Crowd Simulation on Arbitrary Topologies

Recent crowd simulation algorithms do path planning on complex surfaces by breaking 3D surfaces into a series of 2.5D planes. This allows for path planning on surfaces that can be mapped from 3D to 2D without distortion, such as multistory buildings. However, the 2.5D approach does not handle path planning on curved surfaces such as spheres, asteroids, or insect colonies. Additionally, the 2.5D approach does not handle the complexity of dynamic obstacle avoidance when agents can walk on walls or ceilings. We propose novel path planning and obstacle avoidance algorithms that work on surfaces as a whole instead of breaking them into a 2.5D series of planes. Our "whole surfaceâ approach simulates crowds on both multistory structures and highly curved topologies without changing parameters. We validate our work on a suite of 30 different meshes, some with over 100,000 triangles, with crowds of 1,000 agents. Our algorithm always averaged more than 40 FPS with virtually no stalling.

Optimization-based Pedestrian Model Calibration for Evaluation

The evaluation and comparison of crowd simulation algorithms (complex, high-dimensional, multi-scale systems) is an important question. “Realism” being dependent on target applications, comparisons with real measurements are not easy. Promising so- lutions have been suggested for such evaluations (Guy et al. (2012)). Here, we address estimating simulation parameters before evaluating: what do evaluation results mean if the assessed model is not performing at its best? We propose an optimization-based approach encompassing: reference data, metrics, simulation algorithms and optimization techniques. We demonstrate finding good parameter values setting simulation results as close as possible to reference data, enabling fair and meaningful comparisons.

Generating freestyle group formations in agent-based crowd simulations.

Few crowd simulation algorithms take collective group features such as group formations into consideration. Recent approaches often manually specify hard constraints such as predefined keyframe agent formations and interpolate them during the simulation. A proposed interactive, scalable framework generates freestyle group formations and transitions via natural and flexible sketching interaction. It automatically computes a plausible agent distribution in the target formation and agent correspondences between keyframes. In addition, two-level formation trajectory control lets users intuitively guide agents' transition paths from the initial formation to the target formation. In experiments, the framework efficiently generated formations on the fly while providing sufficiently flexible control to users.

Research on Human and Vehicles Coordination Evacuation Simulation Optimization Method under Emergency Incidents

Among researches on crowd flow evacuation under emergency, there are no deep touch of coordinating evacuation strategy of crowd and traffic, and also can not set models and simulate the whole picture of evacuation. The purpose of this research is to realize the coordinating evacuation of both traffic and crowd. The author divided the situations into Not Intervene and May Intervene, analyzed the coordination system and probed the model of coordinating evacuation. A series of coordinate evacuation method has been put forward based on swarm intelligence and simulation optimization. Based on the result of crowd simulation and Ant colony algorithm, in order to find the optimal traffic evacuation proposal, the author took use of the improved Genetic Algorithm as well as the distributed coordinating simulation, finally found a systematic optimal scheduling program. Also the internal regular pattern of coordinating evacuation has also been revealed. All these will provide support for emergency planning and decision making.