Increasing the Efficiency of Policy Learning for Autonomous Vehicles by Multi-Task Representation Learning
Loading...
Access rights
openAccess
URL
Journal Title
Journal ISSN
Volume Title
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
This publication is imported from Aalto University research portal.
View publication in the Research portal (opens in new window)
View/Open full text file from the Research portal (opens in new window)
Other link related to publication (opens in new window)
View publication in the Research portal (opens in new window)
View/Open full text file from the Research portal (opens in new window)
Other link related to publication (opens in new window)
Authors
Date
2022-09
Major/Subject
Mcode
Degree programme
Language
en
Pages
10
701-710
701-710
Series
IEEE Transactions on Intelligent Vehicles, Volume 7, issue 3
Abstract
Driving in a dynamic, multi-agent, and complex urban environment is a difficult task requiring a complex decision-making policy. The learning of such a policy requires a state representation that can encode the entire environment. Mid-level representations that encode a vehicle's environment as images have become a popular choice. Still, they are quite high-dimensional, limiting their use in data-hungry approaches such as reinforcement learning. In this article, we propose to learn a low-dimensional and rich latent representation of the environment by leveraging the knowledge of relevant semantic factors. To do this, we train an encoder-decoder deep neural network to predict multiple application-relevant factors such as the trajectories of other agents and the ego car. Furthermore, we propose a hazard signal based on other vehicles' future trajectories and the planned route which is used in conjunction with the learned latent representation as input to a down-stream policy. We demonstrate that using the multi-head encoder-decoder neural network results in a more informative representation than a standard single-head model. In particular, the proposed representation learning and the hazard signal help reinforcement learning to learn faster, with increased performance and less data than baseline methods.Description
Tallenna OA-artikkeli, kun julkaistu
Keywords
Automobiles, Autonomous vehicles, Hazards, Image reconstruction, Multi-task learning, Policy learning, Reinforcement learning, Representation learning, Task analysis, Trajectory, Vehicle dynamics
Other note
Citation
Kargar, E & Kyrki, V 2022, ' Increasing the Efficiency of Policy Learning for Autonomous Vehicles by Multi-Task Representation Learning ', IEEE Transactions on Intelligent Vehicles, vol. 7, no. 3, pp. 701-710 . https://doi.org/10.1109/TIV.2022.3149891