removed redundant osc2 files

.gitignore

@@ -1,9 +1,5 @@
 *.pyc
-# ignore docker for now
 .ruff_cache
-docker/id_rsa
-id_rsa
-PythonAPI
 srunner/autoagents/*.pyc
 srunner/autoagents/__pycache__/
 srunner/scenariomanager/*.pyc
@@ -13,22 +9,29 @@ srunner/scenarios/__pycache__/
 srunner/utilities/*.pyc
 srunner/utilities/__pycache__/
 results*.txt
-tests
+tests/
 *.log
 .project
 .pydevproject
 *.png
 *.svg
 *.dot
 .python-version
-PythonAPI
-results
+PythonAPI/
+results/
 srunner/osc2
 srunner/osc2_dm
 srunner/osc2_stdlib
 metrics_manager.py
 docker-compose.yml
 jats
 paper
-logs
-typings
+logs/
+typings/
+cawsr_workspace/
+tools/
+scenarios/
+algorithms/
+configs/
+results/
+Dockerfile_dev

CONTRIBUTING.md

@@ -2,6 +2,8 @@
 
 Thank you for your interest in this project! We welcome contributions from the community.
 
+To setup the development environment, please refer to the [docs/DEVELOPMENT.md](docs/DEVELOPMENT.md) file.
+
 ## How to Contribute
 1.  **Fork the repository** on GitHub.
 2.  **Create a new branch** for your feature or bug fix.

Dockerfile

@@ -3,6 +3,10 @@ FROM osrf/ros:humble-desktop
 RUN apt -y update && \
     apt install --no-install-recommends -y libpng16-16 libtiff5 libjpeg8 build-essential curl wget git libxerces-c-dev python3-pip vim
 
+# Create carla user with same UID/GID as the CARLA simulator container
+RUN groupadd -g 1000 carla && \
+    useradd -m -u 1000 -g carla -s /bin/bash carla
+
 # clone repo
 COPY . /autoware_scenario_runner
 
@@ -50,11 +54,19 @@ RUN mkdir /cyclonedds && \
     mv /autoware_scenario_runner/docker/cyclonedds_local.xml /cyclonedds/ && \
     mv /autoware_scenario_runner/docker/cyclonedds_distributed.xml /cyclonedds/ && \
     rm -rf /autoware_scenario_runner/docker && \
-    echo "export RMW_IMPLEMENTATION=rmw_cyclonedds_cpp" >> ~/.bashrc && \
-    echo "alias rossrc='source ${AUTOWARE_MSG_PKG} && source ${ROS_PKG} && echo Sourced'" >> ~/.bashrc && \
-    source ~/.bashrc
+    echo "export RMW_IMPLEMENTATION=rmw_cyclonedds_cpp" >> /home/carla/.bashrc && \
+    echo "alias rossrc='source ${AUTOWARE_MSG_PKG} && source ${ROS_PKG} && echo Sourced'" >> /home/carla/.bashrc
 
 ENV CARLA_API_ROOT="/autoware_scenario_runner/PythonAPI"
 ENV PYTHONPATH="${PYTHONPATH}:${CARLA_API_ROOT}/carla/agents:${CARLA_API_ROOT}/carla"
 
+# Give carla user ownership of necessary directories
+RUN chown -R carla:carla /autoware_scenario_runner && \
+    chown -R carla:carla /ros_workspace && \
+    chown -R carla:carla /cyclonedds && \
+    usermod -aG docker carla
+
+# Switch to carla user
+USER carla
+
 ENTRYPOINT [ "/autoware_scenario_runner/entrypoint.sh" ]

README.md

@@ -1,15 +1,17 @@
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 
+For development and contributing docs see [docs/DEVELOPMENT.md](docs/DEVELOPMENT.md)
+
 CAWSR: ScenarioRunner for CARLA with support for Autoware
 ========================
 
 CAWSR (Carla Autoware Scenario Runner) is a scenario execution engine built for the testing of [Autoware](https://autoware.org/autoware-overview/) in route-based scenarios.
 
 Prerequisites
 ---------------------------
-Both CARLA and Autoware require a high-spec computer with a high-end Nvidia GPU. It is also possible to run a [**distributed**]() setup with multiple machines to help ease the workload, or run the entire stack locally. Currently, only Linux is supported (guide was written on Ubuntu 24.04).
+Both CARLA and Autoware require a high-spec computer with a high-end Nvidia GPU. It is also possible to run a [**distributed**](#distributed) setup with multiple machines to help ease the workload, or run the entire stack locally. Currently, only Linux is supported (guide was written on Ubuntu 24.04).
 
-Ensure the target machine(s) have the [Docker Engine]() and [Nvidia Container toolkit]() installed to enable gpu accelerated workflows in Docker.
+Ensure the target machine(s) have the [Docker Engine](https://docs.docker.com/engine/install/) and [Nvidia Container toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html) installed to enable gpu accelerated workflows in Docker.
 
 CAWSR Setup
 ---------------------------
@@ -45,7 +47,8 @@ Running CAWSR
 
 CAWSR can both be ran `locally` or `distributed`. Due to the high-spec requirements, it is recommended to run distributed if you do not meet the following minimum specs:
 - At least **10GB** VRAM and a modern GPU (2080 ti or newer)
-    - We currently supports 20 and 30 series GPUs. Compatibility with the 40 series is current WIP.     
+    - We currently supports 20 and 30 series GPUs. Compatibility with the 40 series is current WIP.
+
 - At least **32GB** RAM
 - A modern Intel or AMD CPU with at least 8 cores.
 
@@ -117,6 +120,11 @@ docker_compose.yml
 ```
 All folders are mounted as Docker volumes into the CAWSR container, so any changes persist between host and container.
 
+## Autoware Warmup
+
+Before using CAWSR, Autoware must build various `TensorRT` engines for inference. Engine builds are not cross-compatible across various GPU compute capabilities (see [here](https://developer.nvidia.com/cuda/gpus)), so they must be rebuilt per machine. This process usually only takes a few minutes on modern hardware, and starts after the initial launch of autoware.
+
+
 ## Configuring CAWSR
 
 CAWSR is designed to be highly configurable and supports easy swapping of config files.
@@ -167,10 +175,10 @@ We use a custom implementation of a scenario definition in JSON. We have include
 Domain Model:
 ![Domain Model](./docs/resources/scenario_domain.png)
 
-Contributing
-------------
+Contributing & Development
+--------------------------
 
-Please take a look at our [Contribution guidelines]().
+Please take a look at our [Contribution guidelines](CONTRIBUTING.md). To setup a development container, follow our [development guidelines](docs/DEVELOPMENT.md)
 
 License
 -------