Mitigating Covariate Shift in Imitation Learning for Autonomous Vehicles Using Latent Space Generative World Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908344138072064 |
|---|---|
| author | Popov, Alexander Degirmenci, Alperen Wehr, David Hegde, Shashank Oldja, Ryan Kamenev, Alexey Douillard, Bertrand Nistér, David Muller, Urs Bhargava, Ruchi Birchfield, Stan Smolyanskiy, Nikolai |
| author_facet | Popov, Alexander Degirmenci, Alperen Wehr, David Hegde, Shashank Oldja, Ryan Kamenev, Alexey Douillard, Bertrand Nistér, David Muller, Urs Bhargava, Ruchi Birchfield, Stan Smolyanskiy, Nikolai |
| contents | We propose the use of latent space generative world models to address the covariate shift problem in autonomous driving. A world model is a neural network capable of predicting an agent's next state given past states and actions. By leveraging a world model during training, the driving policy effectively mitigates covariate shift without requiring an excessive amount of training data. During end-to-end training, our policy learns how to recover from errors by aligning with states observed in human demonstrations, so that at runtime it can recover from perturbations outside the training distribution. Additionally, we introduce a novel transformer-based perception encoder that employs multi-view cross-attention and a learned scene query. We present qualitative and quantitative results, demonstrating significant improvements upon prior state of the art in closed-loop testing in the CARLA simulator, as well as showing the ability to handle perturbations in both CARLA and NVIDIA's DRIVE Sim. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_16663 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Mitigating Covariate Shift in Imitation Learning for Autonomous Vehicles Using Latent Space Generative World Models Popov, Alexander Degirmenci, Alperen Wehr, David Hegde, Shashank Oldja, Ryan Kamenev, Alexey Douillard, Bertrand Nistér, David Muller, Urs Bhargava, Ruchi Birchfield, Stan Smolyanskiy, Nikolai Robotics Computer Vision and Pattern Recognition Machine Learning Systems and Control 68T40 (Primary) 68T05, 68T45 (Secondary) I.2.9; I.2.6; I.2.10; I.6 We propose the use of latent space generative world models to address the covariate shift problem in autonomous driving. A world model is a neural network capable of predicting an agent's next state given past states and actions. By leveraging a world model during training, the driving policy effectively mitigates covariate shift without requiring an excessive amount of training data. During end-to-end training, our policy learns how to recover from errors by aligning with states observed in human demonstrations, so that at runtime it can recover from perturbations outside the training distribution. Additionally, we introduce a novel transformer-based perception encoder that employs multi-view cross-attention and a learned scene query. We present qualitative and quantitative results, demonstrating significant improvements upon prior state of the art in closed-loop testing in the CARLA simulator, as well as showing the ability to handle perturbations in both CARLA and NVIDIA's DRIVE Sim. |
| title | Mitigating Covariate Shift in Imitation Learning for Autonomous Vehicles Using Latent Space Generative World Models |
| topic | Robotics Computer Vision and Pattern Recognition Machine Learning Systems and Control 68T40 (Primary) 68T05, 68T45 (Secondary) I.2.9; I.2.6; I.2.10; I.6 |
| url | https://arxiv.org/abs/2409.16663 |