Are socially-aware trajectory prediction models

really socially-aware?

Journal of Transportation Research Part C (TR_C), 2022
Given the observation trajectories of the agents in the scene, a predictor (here S-LSTM) forecasts the future positions reasonably (blue lines). However, with less than 5cm perturbation in the observation trajectory (in red), an unacceptable collision is predicted.


Our field has recently witnessed an arms race of neural network-based trajectory predictors. While these predictors are at the core of many applications such as autonomous navigation or pedestrian flow simulations, their adversarial robustness has not been carefully studied. In this paper, we introduce a socially-attended attack to assess the social understanding of prediction models in terms of collision avoidance. An attack is a small yet carefully-crafted perturbations to fail predictors. Technically, we define collision as a failure mode of the output, and propose hard- and soft-attention mechanisms to guide our attack. Thanks to our attack, we shed light on the limitations of the current models in terms of their social understanding. We demonstrate the strengths of our method on the recent trajectory prediction models. Finally, we show that our attack can be employed to increase the social understanding of state-of-the-art models.


title = {Are socially-aware trajectory prediction models really socially-aware?},
journal = {Transportation Research Part C: Emerging Technologies},
volume = {141},
pages = {103705},
year = {2022},
issn = {0968-090X},
doi = {},
author = {Saeed Saadatnejad and Mohammadhossein Bahari and Pedram Khorsandi and Mohammad Saneian and Seyed-Mohsen Moosavi-Dezfooli and Alexandre Alahi},