ll7/carla-rl

carla-rl's Issues

Automate server startup checks

create a pedestrian environment, where the pedestrian is rewarded for walking as far as possible from the spawn point.

Create a gym environment where the pedestrian is rewarded to walk longer distances.

Action space:

• walk direction
• walk speed

observation space

• segmentation camera
• init position
• current position
• current orientation

reward:

only based on the distance from the init position

expected issues:

• action space defintion for a walker
• input layer definition for a camera image and float numbers

Initiation does not set spectator

When the carla server is started for the first time, the spectator is not set to the proper position.
I don't know if the walker is spawned, but it gets deleted.

Synchronous mode causes the actor not to spawn

in synchronous mode the actor does not spawn. Why?

Reintegrate camera observation

improve reward design

Train simple model

create vehicle env

Change vehicle to pedestrian

use an action to create walker behavior

the step function should use a action and a apply a corresponding movement to the walker.

Try to close the loop.

add segmentation camera

Cuda out of memory

RuntimeError: CUDA error: out of memory
CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.

Test Env

observation has an unconventional shape

/home/luttkule/anaconda3/envs/carla-rl-py37-c13/lib/python3.7/site-packages/stable_baselines3/common/env_checker.py:131: UserWarning: Your observation  has an unconventional shape (neither an image, nor a 1D vector). We recommend you to flatten the observation to have only a 1D vector or use a custom policy to properly process the data.
  f"Your observation {key} has an unconventional shape (neither an image, nor a 1D vector). "

change town to something simple like town01

change resolution and field of view of camera

train_vehicle_env

how to solve“Process finished with exit code 139 (interrupted by signal 11: SIGSEGV)”