Coordinated Robot-Assisted Human Crowd Evacuation

Abstract

In this paper, we study the coordinated robot-assisted human crowd evacuation problem, and our aim is to optimally deploy robots and choose proper human-robot interactions to evacuate human crowd in an efficient and safe manner. For such a purpose, we propose a stochastic differential equation model to capture the motion of each individual pedestrian in a crowd. Furthermore, we incorporate the impact of the robots on individual pedestrians in the stochastic differential equation model. From the microscopic pedestrian model, we obtain its corresponding Kolmogorov equation to describe the crowd dynamics evolution at the macroscopic level. The robot deployment and command selection issues are then formulated as an optimal control problem, and a two-step hierarchical structure is proposed to solve the optimal control problem. Simulations are given to illustrate our proposed approach and validate the effectiveness of using multiple robots to assist human crowd evacuation process.