Social navigation tested under varied pedestrian behaviors and crowds
PopNavShift: Stress-Testing Social Navigation under Behavioral Population Shift
Robotics
Summary
Robots that navigate around people usually assume everyone behaves the same way, but people react differently depending on who they are and the situation. The authors created PopNavShift, a simulation tool that changes how pedestrians move based on different personality types and situations to see how well robots handle these changes. They tested three robot navigation styles and found that small changes in pedestrian behavior can significantly affect how efficient and comfortable the robot’s movement is, especially in tight spaces. This work shows that robot navigation should be tested on different types of pedestrian behavior, not just one fixed pattern.
What this means in practice
- •For robotics engineers: Assess robot navigation methods under different pedestrian behaviors to improve real-world robot movement safety and efficiency.
- •For urban planners: Use population-conditioned simulations to evaluate how robots impact pedestrian flow and delays in crowded, constrained spaces.
Authors
Kaizhen Tan, Diyu Zheng, Tim Guangyu Wu, ChengHe Guan
Abstract
Social-navigation algorithms are often evaluated under a fixed pedestrian-behavior distribution, despite substantial variation in pedestrian responses to robots across individuals and social contexts. We introduce PopNavShift, a matched simulation framework for stress-testing social-navigation strategies under pedestrian population shifts. PopNavShift constructs population-conditioned pedestrian motion profiles by prompting Gemini 3.7 Flash with 600 synthetic persona records from MatrAIx Persona 1M and deterministically mapping the responses into bounded motion parameters. It then compares three representative navigation strategies, reactive avoidance, early yielding, and reciprocal collision avoidance, across eight population conditions and 7,488 matched robot runs. In a matched intervention on the same 202 personas, changing only time pressure reverses 8.6% of controller rankings based on robot travel time, but 22.4% based on mean pedestrian delay and 23.9% based on worst-decile delay. Across population conditions, this sensitivity is greater for pedestrian burden than for robot travel time and increases in spatially constrained settings; the same qualitative pattern persists under a second pedestrian dynamics model. These findings support evaluating navigation strategies across behavioral populations using both robot performance and pedestrian burden.