论文信息 - Inverse Reinforcement Learning with BDI Agents for Pedestrian Behavior Simulation

Inverse Reinforcement Learning with BDI Agents for Pedestrian Behavior Simulation

Crowd behavior has been subject of study in fields like disaster evacuation, smart town planning and business strategic placing. It is possible to create a model for those scenarios using machine learning techniques and a relatively small training data set to identify behavioral. We implemented a BDI-based agent model that uses such techniques into a large-scale crowd simulator, and apply inverse reinforcement learning to adjust agents' behaviors by examples. The goal of the system is to provide to the agents a realistic behavior model and a method to orient themselves without knowing the scenario's layout, based in learnt patterns around environment features.

Nahum Alvarez

[1] Heinz Koeppl,et al. Inverse Reinforcement Learning in Swarm Systems , 2016, AAMAS.

[2] Paul M. Torrens,et al. An extensible simulation environment and movement metrics for testing walking behavior in agent-based models , 2012, Comput. Environ. Urban Syst..

[3] Stefania Bandini,et al. Route choice in pedestrian simulation: Design and evaluation of a model based on empirical observations , 2016, Intelligenza Artificiale.

[4] Ana L. C. Bazzan,et al. Understanding the Behaviour of Learning-Based BDI Agents in the Braess' Paradox , 2017, MATES.

[5] Andrew Y. Ng,et al. Pharmacokinetics of a novel formulation of ivermectin after administration to goats , 2000, ICML.

[6] Pieter Abbeel,et al. Apprenticeship learning via inverse reinforcement learning , 2004, ICML.

[7] Fernando Fernández,et al. Emergent behaviors and scalability for multi-agent reinforcement learning-based pedestrian models , 2017, Simul. Model. Pract. Theory.

[8] Wentong Cai,et al. Agent‐based human behavior modeling for crowd simulation , 2008, Comput. Animat. Virtual Worlds.

[9] Itsuki Noda,et al. Evacuation Planning Assist System with Network Model-Based Pedestrian Simulator , 2009, PRIMA.