Abstract:For walking pedestrians, when they are blocked by obstacles or other pedestrians, they adjust their speeds and directions to avoid colliding with them, which is called collision avoidance behavior. This behavior is the most complex part of pedestrians' walking processes and its modeling and simulation are the keys to realistic crowd simulation, which serves as the foundation for various applications. However, most existing methods either lack the representation power to accurately model the complex collision behavior or do not model it explicitly, which leads to a poor level of realism of the simulation. To realize realistic crowd simulation, we propose to analyze, understand, and model the collision avoidance behavior in a data-driven way. First, to automatically detect collision avoidance behavior for further analysis, we propose a domain transformation algorithm that detects it by transforming the trajectories in the spatial domain into a new domain where the behavior is much more apparent and is thus easier to detect. The new domain also provides a new perspective for understanding collision avoidance behavior. Second, since there are no mature metrics to evaluate the level of realism, we propose a new evaluation metric based on the least-effort theory, which evaluates the realism of collision avoidance behavior by its physical and mental consumption. This evaluation metric also provides the foundation of modeling. Third, for realistic crowd simulation, we design a reinforcement learning model. It trains agents with our proposed reward function that models pedestrians' intrinsic needs of "reducing effort consumption'' and thus can guide agents to behave realistically when avoiding collisions. Extensive experiments show our model is 55.9% and 52.5% more realistic in collision avoidance behavior than the best baselines on two real-world datasets. We release our codes at https://github.com/tsinghua-fib-lab/TECRL.

A Data-Driven Approach for Simulating Pedestrian Collision Avoidance in Crossroads

A New Collision Avoidance Model for Pedestrian Dynamics

A Data Driven Simulation Model for Investigating Collision Avoidance Behavior of Pedestrians in Subway Stations

Understanding and Modeling Collision Avoidance Behavior for Realistic Crowd Simulation

A modified social force model considering collision avoidance based on empirical studies

Deconstructing Pedestrian Crossing Decision-making in Interactions with Continuous Traffic: an Anthropomorphic Model

Simulation of Pedestrian Crossing Behaviors at Unmarked Roadways Based on Social Force Model

A Cellular Automaton Simulation Model for Pedestrian and Vehicle Interaction Behaviors at Unsignalized Mid-Block Crosswalks.

Modeling and Validating Pedestrian Dynamics at Crosswalk: A Social-Forced Based Approach

Simulation of Pedestrian Behavior in the Collision-Avoidance Process Considering Their Moving Preferences

Development of pedestrian collision avoidance strategy based on the fusion of Markov and social force models

A Modified Heuristics-Based Model for Simulating Realistic Pedestrian Movement Behavior

Pedestrian Interaction Algorithm for Autonomous Vehicles Considering Crossing Intention and Driving Style

Deconstructing Pedestrian Crossing Decisions in Interactions with Continuous Traffic: an Anthropomorphic Model

Experimental study on pedestrians' collision avoidance

An Intelligent Pedestrian-Vehicle Collision Avoidance Strategy Based on Evolutionary Game

A collision-free model on the interaction between pedestrians and cyclists on a shared road

Less-Effort Collision Avoidance in Virtual Pedestrian Simulation

Data-Driven and Collision-Free Hybrid Crowd Simulation Model for Real Scenario

A Data-Driven Neural Network Approach to Simulate Pedestrian Movement

Decision Model for Pedestrian Interacting with Traffic at Uncontrolled Intersections