Visual-based Autonomous Driving Deployment from a Stochastic and Uncertainty-aware Perspective
This work addresses safety issues in autonomous driving deployment by reducing uncertainty, but it is incremental as it builds on existing real-to-sim pipelines and imitation learning methods.
The paper tackles the problem of deploying visual-based autonomous driving policies by addressing input uncertainty, using a stochastic generator to translate test images to the training domain and an uncertainty-aware policy to select safe actions. Experiments on the Carla benchmark show it outperforms previous methods, particularly in dynamic environments.
End-to-end visual-based imitation learning has been widely applied in autonomous driving. When deploying the trained visual-based driving policy, a deterministic command is usually directly applied without considering the uncertainty of the input data. Such kind of policies may bring dramatical damage when applied in the real world. In this paper, we follow the recent real-to-sim pipeline by translating the testing world image back to the training domain when using the trained policy. In the translating process, a stochastic generator is used to generate various images stylized under the training domain randomly or directionally. Based on those translated images, the trained uncertainty-aware imitation learning policy would output both the predicted action and the data uncertainty motivated by the aleatoric loss function. Through the uncertainty-aware imitation learning policy, we can easily choose the safest one with the lowest uncertainty among the generated images. Experiments in the Carla navigation benchmark show that our strategy outperforms previous methods, especially in dynamic environments.