mlsanityData to 14 Oct 2026

Robot Navigation in Crowds by Graph Convolutional Networks with Attention Learned from Human Gaze

arXiv 1909.10400 ↗First posted 23 Sep 2019cs.RO cs.AI cs.CVcited by 30

Not on the map. Why?

Abstract (arXiv)

Safe and efficient crowd navigation for mobile robot is a crucial yet challenging task. Previous work has shown the power of deep reinforcement learning frameworks to train efficient policies. However, their performance deteriorates when the crowd size grows. We suggest that this can be addressed by enabling the network to identify and pay attention to the humans in the crowd that are most critical to navigation. We propose a novel network utilizing a graph representation to learn the policy. We first train a graph convolutional network based on human gaze data that accurately predicts human attention to different agents in the crowd. Then we incorporate the learned attention into a graph-based reinforcement learning architecture. The proposed attention mechanism enables the assignment of meaningful weightings to the neighbors of the robot, and has the additional benefit of interpretability. Experiments on real-world dense pedestrian datasets with various crowd sizes demonstrate that our model outperforms state-of-art methods by 18.4% in task accomplishment and by 16.4% in time efficiency.

Key facts. Cited by 30 papers in the map (data through 14 Oct 2026). First posted on arXiv 23 Sep 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2020: 2; 2021: 7; 2022: 5; 2023: 4; 2024: 6; 2025: 3; 2026: 3.

30 papers.

Sort:

By area:

  1. 2022Intention Aware Robot Crowd Navigation with Attention-Based Interaction Graphcited by 53
  2. 2020Towards Target-Driven Visual Navigation in Indoor Scenes via Generative Imitation Learningcited by 14
  3. 2021Recent Advances in Leveraging Human Guidance for Sequential Decision-Making Taskscited by 13
  4. 2021DQ-GAT: Towards Safe and Efficient Autonomous Driving with Deep Q-Learning and Graph Attention Networkscited by 13
  5. 2021AVGCN: Trajectory Prediction using Graph Convolutional Networks Guided by Human Attentioncited by 12
  6. 2024DR-MPC: Deep Residual Model Predictive Control for Real-world Social Navigationcited by 9
  7. 2024HEIGHT: Heterogeneous Interaction Graph Transformer for Robot Navigation in Crowded and Constrained Environmentscited by 9
  8. 2020Selective Eye-gaze Augmentation To Enhance Imitation Learning In Atari Gamescited by 6
  9. 2021Robot Navigation in a Crowd by Integrating Deep Reinforcement Learning and Online Planningcited by 6
  10. 2023TBD Pedestrian Data Collection: Towards Rich, Portable, and Large-Scale Natural Pedestrian Datacited by 6
  11. 2024Socially Integrated Navigation: A Social Acting Robot with Deep Reinforcement Learningcited by 5
  12. 2024Attention Graph for Multi-Robot Social Navigation with Deep Reinforcement Learningcited by 4
18 more
  1. 2021Autonomous Navigation through intersections with Graph ConvolutionalNetworks and Conditional Imitation Learning for Self-driving Carscited by 3
  2. 2021A Graph Neural Network to Model Disruption in Human-Aware Robot Navigationcited by 3
  3. 2024HeR-DRL:Heterogeneous Relational Deep Reinforcement Learning for Decentralized Multi-Robot Crowd Navigationcited by 3
  4. 2025COLSON: Controllable Learning-Based Social Navigation via Diffusion-Based Reinforcement Learningcited by 3
  5. 2025Socially aware navigation for mobile robots: a survey on deep reinforcement learning approachescited by 2
  6. 2022Multi-subgoal Robot Navigation in Crowds with History Information and Interactionscited by 1
  7. 2022Dense Crowd Flow-Informed Path Planningcited by 1
  8. 2022Active Classification of Moving Targets with Learned Control Policiescited by 1
  9. 2023SOCRATES: Text-based Human Search and Approach using a Robot Dogcited by 1
  10. 2023InCrowdFormer: On-Ground Pedestrian World Model From Egocentric Viewscited by 1
  11. 2024MeSA-DRL: Memory-Enhanced Deep Reinforcement Learning for Advanced Socially Aware Robot Navigation in Crowded Environmentscited by 1
  12. 2026Incremental Residual Reinforcement Learning Toward Real-World Learning for Social Navigationcited by 1
  13. 2021Relative velocity-based reward functions for crowd navigation of robotscited by 0
  14. 2022Towards Rich, Portable, and Large-Scale Pedestrian Data Collectioncited by 0
  15. 2023Semantic and Articulated Pedestrian Sensing Onboard a Moving Vehiclecited by 0
  16. 2025Integrating Offline Pre-Training with Online Fine-Tuning: A Reinforcement Learning Approach for Robot Social Navigationcited by 0
  17. 2026Learning-Based Motion Planning for Dynamic Environments: From Foundational Algorithms to Emerging Paradigmscited by 0
  18. 2026Performance-Preserving Online Adaptation in Social Navigation via Diffusion Steeringcited by 0