exploring_limitations.md

Exploring the Limitations of Behavior Cloning for Autonomous Driving

Downloading Checkpoints

The configs/nocrash folder has some of the relevant models that can be found on the paper.

The following script will download all the needed files.

python3 tools/download_nocrash_models.py

Models available

• resnet34imnet10S1: is the model with the random seed 1 from Figure 6 .
• resnet34imnet10S2: is the model with the random seed 2 from Figure 6 and it is also our best model (Green Model Fig. 5).
• resnet34imnet10-nospeed: without the speed prediction and ten hours of training (Yellow model Fig. 5).
• resnet34imnet100: the model with 100 hours of demonstrations (Blue model Fig. 5).
• resnet34imnet100-nospeed: the model with 100 hours of demonstrations and no speed prediction (Red model Fig. 5).

Reproducing the results

To reproduce all of the available models, using the gpu 0, run:

python3 coiltraine.py --gpus 0 --folder nocrash -de NocrashNewWeatherTown_Town02 NocrashNewWeather_Town01\
 NocrashTraining_Town01 NocrashNewTown_Town02 --docker carlagear

Note, the models use the CARLA single gear version of 0.8.4. This is discussed on this repository:

https://github.com/carla-simulator/data-collector

To test our best model on the hardest condition:

python3 coiltraine.py --gpus 0 --single-process drive -e resnet34imnet10S2 --folder nocrash \
-de NocrashNewWeatherTown_Town02 --docker carlagear

If you use any of our baselines please cite our paper:

Add Soon