Find your next open-source project

Search by name, purpose, or keyword — or paste a GitHub repository URL
All skills

carla-simulator/carla

Installs and runs CARLA, the open-source autonomous-driving simulator, covering setup, maps, vehicles, sensors, scenarios and ROS 2.

  • 69 skills
  • 14.6k stars

The 5 most installed

  • Adds a new ROS 2 message type to CARLA's native ROS 2 interface, including serialization and type hash.

  • Adds a new ROS 2 topic, publisher or subscriber to CARLA's native ROS 2 interface.

  • Reads ScenarioRunner output, summarises pass/fail criteria and computes custom metrics over recorded runs offline.

  • Finds undocumented features, stale API references, broken links and unreachable pages in a CARLA checkout.

  • Gets actor and level 3D bounding boxes and projects them into camera images as 2D annotations.

The other 64 skills
  • Builds CARLA on Unreal Engine 4.26 from source on Linux, including the server, Python client and content.

  • Builds CARLA on Unreal Engine 5.8 using its CMake presets and targets.

  • Creates, migrates and debugs detailed vehicle assets in Blender and Unreal for CARLA.

  • Reports which network ports CARLA and its tools use and which are already taken.

  • Tells you whether CARLA supports a given feature on this build and detects version mismatches.

  • Lists what CARLA on Unreal Engine 5.5 cannot do compared with 5.8, so you avoid unsupported workflows.

  • Turns street, building and vehicle lights on or off and sets their colour, intensity and night cycle.

  • Moves the spectator camera, frames an actor or follows a vehicle in third-person, first-person or top-down view.

  • Tunes the Traffic Manager that drives autopilot cars: speed, following distance, lane changes, lights and determinism.

  • Sets traffic lights green, red or yellow, freezes them, changes phase timing or resets the normal cycle.

  • Drives a specific vehicle directly with throttle, steering, brake, lights, doors and physics tuning.

  • Builds a CARLA content pack of maps, props, vehicles or walkers that installs into a UE 5.8 package.

  • Creates, edits, visualises and summarises CARLA Leaderboard route files with waypoints and scenario triggers.

  • Writes and registers a new ScenarioRunner scenario class with its XML config, validating it without a simulator.

  • Turns a natural-language description into a Scenic scenario file and validates it before running.

  • Spawns and attaches cameras, lidar, radar, IMU, GNSS and other sensors to actors, with ROS 2 publishing.

  • Draws debug points, lines, arrows, boxes and text labels in the CARLA world.

  • Visualises and validates a map's pedestrian navigation mesh by sampling and drawing walkable locations.

  • Writes and reviews CARLA documentation in house style, checking every claim against the code first.

  • Downloads a CARLA release package, nightly build, source checkout or Docker image and reports what was obtained.

  • Downloads or builds the AdditionalMaps pack and installs it into a CARLA UE 5.8 package after checking it matches.

  • Imports a custom RoadRunner map into a CARLA source build and verifies it loads on a server.

  • Imports FBX meshes into a CARLA source build as spawnable static props and verifies them by spawning.

  • Explains and runs the ways to get a custom map into CARLA on UE 5.8, from build-time import to runtime OpenDRIVE loading.

  • Imports an FBX mesh into CARLA on UE 5.8 as a spawnable prop and registers it in the blueprint library.

  • Imports a skinned pedestrian model into CARLA on UE 5.8 as a spawnable, animating walker.

  • Imports an FBX vehicle into a CARLA source build, assembles it and verifies it spawns, drives and steers.

  • Imports an FBX pedestrian into a CARLA source build and verifies it spawns, stands and walks.

  • Installs, lists, replaces or removes CARLA content packs in a UE 5.8 package after checking compatibility.

  • Installs a chosen CARLA Leaderboard version with its matching ScenarioRunner and CARLA build, then verifies the stack.

  • Installs the CARLA Python client into a chosen interpreter and verifies it matches the simulator version.

  • Installs CARLA's ScenarioRunner on the branch matching your CARLA version and verifies the import.

  • Installs Scenic, the scenario description language, into the same interpreter as the CARLA client and verifies it.

  • Fills a running CARLA server with Inverted AI model-driven NPC traffic instead of the rule-based Traffic Manager.

  • Loads, reloads or generates maps on a running CARLA server and manages map layers through the Python API.

  • Lists, mounts and unmounts content packs on a running CARLA UE 5.8 server without restarting it.

  • Explores a CARLA map's lanes, junctions and topology and helps locate or navigate specific road elements.

  • Drives a CARLA vehicle autonomously to a destination or plans a route between two points.

  • Takes a CARLA contribution from bug report or feature request through branch, diff review and pull request.

  • Cooks and packages CarlaUE4 into distributable tarballs, either the full simulator or selected maps and props.

  • Cooks and packages a shippable CARLA server from a UE 5.8 checkout and inspects the package for hidden failures.

  • Builds and verifies the Docker image for submitting an agent to the CARLA Leaderboard.

  • Inspects a recorded CARLA log to show its contents, collisions and vehicles that got stuck.

  • Reads a CARLA Leaderboard results file, explains the score, rescores under other rules and diffs runs.

  • Listens to a CARLA sensor and saves, displays or summarises its data, or reports its ROS 2 topics.

  • Records a running CARLA simulation to a log file so you can replay the same run later.

  • Replays a recorded CARLA run, whole or partial, at any speed and following a chosen vehicle.

  • Runs the Autoware self-driving stack against CARLA on UE 5.8 and sets up the maps it needs.

  • Starts a CARLA server from a release, package or source checkout, headless or windowed, and stops it cleanly.

  • Starts a CARLA UE 5.8 server in editor, packaged or headless mode with custom ports and ROS 2.

  • Runs the CARLA Leaderboard evaluator over chosen routes with your agent and checks the setup first.

  • Runs ASAM OpenSCENARIO 1.x and 2.0 scenario files through ScenarioRunner and validates them first.

  • Runs a route through ScenarioRunner's route mode with an agent driving, including route-only scenarios.

  • Runs a single ScenarioRunner Python scenario or a group against a running CARLA server.

  • Runs Scenic probabilistic traffic scenarios against a running CARLA server and restores the world after.

  • Reads and sets CARLA weather, turning plain requests like heavy rain at sunset into exact settings.

  • Switches a CARLA server between sync and async mode and sets fixed timestep, substepping and rendering.

  • Spawns vehicles on a CARLA map, scattered on autopilot, in a lane row, or as the ego car.

  • Spawns pedestrians that wander the map on their own and removes them in the correct order.

  • Reads one actor's live position, speed, acceleration and control input from a running CARLA server.

  • Lists and hides or shows a CARLA map's buildings, trees, signs and other baked environment objects.

  • Checks and visualises a ROS 2-enabled CARLA server's topics and opens RViz from Docker.

  • Finds actors in the CARLA world by type, role or colour and reads the current world snapshot.

  • Writes a CARLA Leaderboard agent with sensors and run steps, checking the sensor budget offline.