Crowd Navigation

Crowd navigation focuses on enabling robots to safely and efficiently traverse environments populated by humans, prioritizing both collision avoidance and socially acceptable behavior. Current research emphasizes developing robust and computationally efficient algorithms, often employing deep reinforcement learning (DRL), model predictive control (MPC), and graph neural networks (GNNs) to predict pedestrian trajectories and plan optimal robot paths. These advancements are crucial for improving the safety and usability of autonomous robots in shared spaces, with applications ranging from service robotics to autonomous vehicles.