Social Attention: Modeling Attention in Human Crowds

Vemula, Anirudh; Muelling, Katharina; Oh, Jean

doi:10.48550/arxiv.1710.04689

Cited by 6 publications

(9 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Yet, the generative models still suffer from mode collapse [42]. Some other works model the social interactions through spatiotemporal graphs, where an attention model is introduced recently to learn the relative importance of each agent [14]. Sadeghian et al [43] study various attention mechanisms.…”

Section: Background a Related Workmentioning

confidence: 99%

Crowd-Robot Interaction: Crowd-Aware Robot Navigation With Attention-Based Deep Reinforcement Learning

Chen

Liu

Kreiss

et al. 2019

2019 International Conference on Robotics and Automation (ICRA)

428

354

View full text Add to dashboard Cite

This paper proposes an end-to-end deep reinforcement learning approach for mobile robot navigation with dynamic obstacles avoidance. Using experience collected in a simulation environment, a convolutional neural network (CNN) is trained to predict proper steering actions of a robot from its egocentric local occupancy maps, which accommodate various sensors and fusion algorithms. The trained neural network is then transferred and executed on a real-world mobile robot to guide its local path planning. The new approach is evaluated both qualitatively and quantitatively in simulation and realworld robot experiments. The results show that the map-based end-to-end navigation model is easy to be deployed to a robotic platform, robust to sensor noise and outperforms other existing DRL-based models in many indicators.

show abstract

Section: Background a Related Workmentioning

confidence: 99%

Crowd-Robot Interaction: Crowd-Aware Robot Navigation With Attention-Based Deep Reinforcement Learning

Chen

Liu

Kreiss

et al. 2019

2019 International Conference on Robotics and Automation (ICRA)

428

354

View full text Add to dashboard Cite

show abstract

“…More recently, a variety of neural networks have been explored for learning socially-aware motion representations [5,43]. Some peculiar neural architecture designs, such as feature pooling [3,21,27], attention mechanism [12,34,75,86], and spatial-temporal graph [36,41,50,81], have yielded promising results in crowded environments. However, the robustness of these methods remains a central concern.…”

Section: Related Workmentioning

confidence: 99%

“…However, building predictive models capable of doing so is challenging. Recent works have proposed a plethora of neural network-based models [21,34,36,41,50,75,81,86] to learn socially-aware motion representations and demonstrated their potentials for human trajectory forecasting [3,27,47,76] or robot motion planning [12,14,15] in crowded spaces. Yet existing methods still output unacceptable solutions (e.g., collisions), which raises significant safety concerns.…”

Section: Introductionmentioning

confidence: 99%

Social NCE: Contrastive Learning of Socially-aware Motion Representations

Liu¹,

Yan²,

Alahi³

2020

Preprint

View full text Add to dashboard Cite

Learning socially-aware motion representations is at the core of recent advances in human trajectory forecasting and robot navigation in crowded spaces. Yet existing methods often struggle to generalize to challenging scenarios and even output unacceptable solutions (e.g., collisions). In this work, we propose to address this issue via contrastive learning. Concretely, we introduce a social contrastive loss that encourages the encoded motion representation to preserve sufficient information for distinguishing a positive future event from a set of negative ones. We explicitly draw these negative samples based on our domain knowledge about socially unfavorable scenarios in the multi-agent context. Experimental results show that the proposed method consistently boosts the performance of previous trajectory forecasting, behavioral cloning, and reinforcement learning algorithms in various settings. Our method makes little assumptions about neural architecture designs, and hence can be used as a generic way to incorporate negative data augmentation into motion representation learning.

show abstract

“…All of these neural models were predicting the trajectory of individual agents without considering any interactions between members of a crowd. Social-LSTM [4] introduced agent interactions in a RNN model of trajectory prediction, and has been extended in Social Attention [8], which has demonstrated improved accuracy to both Vannilla RNN models and Social-LSTM. Social Attention makes use of Structural RNNs [7], which represent the model as a spatio-temporal graph, and allow the modelling of interactions between various agent types.…”

Section: B Trajectory Predictionmentioning

confidence: 99%

“…Nodes: For each node v ∈ V corresponding to a noncontrolled agent, we use an attention module in the same manner as [8] to determine the inputs from each neighbouring node (Eqn. 4).…”

Section: B Model Architecturementioning

confidence: 99%

Predicting Responses to a Robot's Future Motion using Generative Recurrent Neural Networks

Eiffert¹,

Sukkarieh²

2019

Preprint

View full text Add to dashboard Cite

Robotic navigation through crowds or herds requires the ability to both predict the future motion of nearby individuals and understand how these predictions might change in response to a robot's future action. State of the art trajectory prediction models using Recurrent Neural Networks (RNNs) do not currently account for a planned future action of a robot, and so cannot predict how an individual will move in response to a robot's planned path. We propose an approach that adapts RNNs to use a robot's next planned action as an input alongside the current position of nearby individuals. This allows the model to learn the response of individuals with regards to a robot's motion from real world observations. By linking a robot's actions to the response of those around it in training, we show that we are able to not only improve prediction accuracy in close range interactions, but also to predict the likely response of surrounding individuals to simulated actions. This allows the use of the model to simulate state transitions, without requiring any assumptions on agent interaction. We apply this model to varied datasets, including crowds of pedestrians interacting with vehicles and bicycles, and livestock interacting with a robotic vehicle.

show abstract

Social Attention: Modeling Attention in Human Crowds

Cited by 6 publications

References 0 publications

Crowd-Robot Interaction: Crowd-Aware Robot Navigation With Attention-Based Deep Reinforcement Learning

Crowd-Robot Interaction: Crowd-Aware Robot Navigation With Attention-Based Deep Reinforcement Learning

Social NCE: Contrastive Learning of Socially-aware Motion Representations

Predicting Responses to a Robot's Future Motion using Generative Recurrent Neural Networks

Contact Info

Product

Resources

About