Title | ||
---|---|---|
Increasing the Efficiency of Policy Learning for Autonomous Vehicles by Multi-Task Representation Learning |
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. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1109/TIV.2022.3149891 | IEEE Transactions on Intelligent Vehicles |
Keywords | DocType | Volume |
Autonomous vehicles,representation learning,policy learning,multi-task learning | Journal | 7 |
Issue | ISSN | Citations |
3 | 2379-8858 | 0 |
PageRank | References | Authors |
0.34 | 5 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Eshagh Kargar | 1 | 0 | 0.34 |
V. Kyrki | 2 | 652 | 61.79 |