Minimally-Intrusive Navigation in Densely-Populated Pedestrian Flow
Project Description
-
This project aims to address the "freezing robot problem," which refers to the situation where a robot, when faced with a level of environmental complexity (such as pedestrian interaction models) beyond a certain threshold, perceives no path as safe and consequently chooses to remain stationary.
-
This project tackles the issue by defining two levels of disturbances.
-
This project has demonstrated the effectiveness of the algorithm through both simulation experiments and real-world validations.
-
More information can be found in paper 1 and paper2 .
-
This project is still being updated, and you are welcome to check back anytime.
Project Overall
-
From the macroscopic perspective, we quantify the disturbance as the overall impact on pedestrian flow caused by the robot's presence.
-
On the microscopic level, the disturbance is defined as the discrepancy between the future trajectory distributions of individual pedestrians with and without the robot.
Visualization of microscopic level disturbance
Visualization of macro level disturbance
Project Simulation Demonstration
( Real World Experiment is on the top of the page)
