Implementation:CARLA simulator Carla World Apply Settings

Knowledge Sources	CARLA Simulator CARLA Documentation
Domains	Autonomous Driving Simulation, Real-Time Systems
Last Updated	2026-02-15 00:00 GMT

Overview

Concrete tool for applying simulation configuration parameters to the CARLA world provided by the CARLA simulator.

Description

The World.apply_settings method sends a WorldSettings object to the server, updating the global simulation parameters. This includes toggling synchronous mode, setting a fixed time step, enabling or disabling rendering, and configuring physics substepping. The method returns the frame ID at which the new settings take effect, allowing the client to synchronize its state.

The WorldSettings object acts as a configuration bundle that encapsulates all tunable simulation parameters. Changes are atomic -- the entire settings object is applied at once, preventing inconsistent intermediate states.

Usage

Use this method at the start of a simulation session to configure the desired execution mode, or at runtime to dynamically switch between synchronous and asynchronous operation. It is common practice to store the original settings before modification and restore them during cleanup.

Code Reference

Source Location

• Repository: CARLA
• File: LibCarla/source/carla/client/World.cpp
• Lines: L56-84
• Python binding: PythonAPI/carla/src/World.cpp
• Lines: L303-306

Signature

World.apply_settings(settings: carla.WorldSettings, timeout: datetime.timedelta = 0) -> int

Import

import carla

I/O Contract

Inputs

Name	Type	Required	Description
settings	carla.WorldSettings	Yes	Configuration object containing simulation parameters such as synchronous_mode, fixed_delta_seconds, no_rendering_mode, substepping, max_substep_delta_time, and max_substeps
timeout	datetime.timedelta	No	Maximum time to wait for the server to acknowledge the settings change. Defaults to 0 (use the client default timeout).

Outputs

Name	Type	Description
return	int	The frame ID at which the new settings take effect

Usage Examples

Basic Example

import carla

# Connect and get world
client = carla.Client("localhost", 2000)
client.set_timeout(10.0)
world = client.get_world()

# Save original settings for later restoration
original_settings = world.get_settings()

# Configure synchronous mode with fixed time step
settings = world.get_settings()
settings.synchronous_mode = True
settings.fixed_delta_seconds = 0.05  # 20 FPS simulation rate

# Apply settings and get the frame ID
frame_id = world.apply_settings(settings)
print(f"Settings applied at frame: {frame_id}")

# Run simulation loop
try:
    for i in range(100):
        world.tick()  # Advance simulation by one fixed step
finally:
    # Restore original settings on cleanup
    world.apply_settings(original_settings)

High-Throughput Training Example

import carla

client = carla.Client("localhost", 2000)
client.set_timeout(10.0)
world = client.get_world()

original_settings = world.get_settings()

# Configure for maximum throughput: synchronous, no rendering, large time step
settings = world.get_settings()
settings.synchronous_mode = True
settings.fixed_delta_seconds = 0.1       # 10 FPS (fewer ticks per second of sim time)
settings.no_rendering_mode = True         # Disable GPU rendering
settings.substepping = True
settings.max_substep_delta_time = 0.01
settings.max_substeps = 10

world.apply_settings(settings)

try:
    # Training loop runs much faster without rendering overhead
    for episode in range(1000):
        snapshot = world.tick()
        # ... collect observations, compute rewards, apply actions ...
finally:
    world.apply_settings(original_settings)

Related Pages

Implements Principle

• Principle:CARLA_simulator_Carla_Simulation_Configuration

Uses Heuristic

• Heuristic:CARLA_simulator_Carla_Synchronous_Mode_Fixed_Delta