It is reasonable to expect that robots act in a way that does not hinder humans present in the environment. Previous attempts to solve the problem of social navigation either lack flexibility or needs a large volume of data. A recent paper suggests tackling current limitations.
Image credit: Richard Greenhill and Hugo Elias/Wikipedia/CC BY-SA 3.0
The researchers propose a virtual environment that allows transforming the abundant bird’s-eye view data to first-person views of all agents present in a given scene. A novel supervised learning model imitates the social behaviors of real humans in crowded environments using a first-person depth view.
In order to imitate the social behaviors of real humans, it learns to replicate the navigational patterns of the agents present in the data. The results show that the suggested model can learn and then infer rich information about the intentions and potential trajectories of other passers-by and therefore outperforms the baselines.
Current datasets to train social behaviors are usually borrowed from surveillance applications that capture visual data from a bird’s-eye perspective. This leaves aside precious relationships and visual cues that could be captured through a first-person view of a scene. In this work, we propose a strategy to exploit the power of current game engines, such as Unity, to transform pre-existing bird’s-eye view datasets into a first-person view, in particular, a depth view. Using this strategy, we are able to generate large volumes of synthetic data that can be used to pre-train a social navigation model. To test our ideas, we present DeepSocNav, a deep learning based model that takes advantage of the proposed approach to generate synthetic data. Furthermore, DeepSocNav includes a self-supervised strategy that is included as an auxiliary task. This consists of predicting the next depth frame that the agent will face. Our experiments show the benefits of the proposed model that is able to outperform relevant baselines in terms of social navigation scores.
Research paper: de Vicente, J. P. and Soto, A., “DeepSocNav: Social Navigation by Imitating Human Behaviors”, 2021. Link: https://arxiv.org/abs/2107.09170