Configuring the Simulator

The configuration of the Simulator's state is done through the use of YAML files. Below is an example of a sample YAML file.

robots:
- id: robot0
  pose:
  - -6.295703
  - 7.076855
  - 0.0
  vel:
  - 0.0
  - 0.0
  lidar: true
  footprint:
  - 0.25
  drive_type: Differential
  add_noise: false
- id: robot1
  pose:
  - -4.0486326
  - 4.9811525
  - 0.0
  vel:
  - 0.0
  - 0.0
  lidar: true
  footprint:
  - 0.25
  drive_type: Differential
  add_noise: false
walls:
- endpoints:
  - - -9.1560545
    - 8.704004
  - - -1.0940428
    - 8.6789055
  - - -0.99091816
    - 3.9499025
  - - -9.038574
    - 4.1187496
- endpoints:
  - - 1.0652351
    - 8.664355
  - - 1.0892582
    - 4.006152
  - - 9.079687
    - 4.0520506
static_objects:
- center:
  - 2.2917004
  - 6.981738
  width: 0.625
  height: 0.625
  rotation: 0.0
- center:
  - 7.0378914
  - 6.6941404
  width: 0.625
  height: 0.625
  rotation: 0.0
- center:
  - 0.020019531
  - 2.6319332
  width: 0.625
  height: 0.625
  rotation: 0.0

Three categories of objects

There are three main categories of Objects used in the Simulator, Robot, Static Object, Wall. The configuration for each of these is described differently.

In general, the YAML can be split as a combination of these categories as below

robots:
    - ...
    - ...
    - ...
walls:
    - ...
    - ...
    - ...
static_objects:
    - ...
    - ...
    - ...

Configuring a Robot

Property	Description	Data Type
id	Describes the id of the Robot.	string
pose	Describes the pose of the Robot. This is a tuple of [x, y, yaw].	Tuple of [x, y, yaw]
vel	Describes the velocity of the Robot. Varies according to drive type: Differential drive: [vx, vy] Omnidirectional drive: velocity in x and y direction Ackermann drive: linear velocity and steering angle.	Varies according to drive type: Differential drive: [vx, vy] Omnidirectional drive: velocity in x and y direction Ackermann drive: linear velocity and steering angle.
lidar	Describes the presence of a lidar.	bool
footprint	Describes the footprint. If float, the shape of the robot is circular with the given number as radius. If a tuple of float, the first number is taken as width and the second number as height.	float or Tuple of float
drive_type	Describes the drive type. Can be Differential, OmniDirectional, or Ackermann.	string
add_noise	Adds noise to the kinematics model.	bool

An example YAML configuration with Ackermann drive and Rectangular Footprint is given below

- id: robot1
  pose:
    - -4.825901
    - 8.391234
    - 0.35
  vel:
    - 1.5
    - 0.0
  lidar: true
  footprint:
    - 0.4
    - 1.2
  drive_type: Ackermann
  add_noise: true

Configuring a Static Object

Here's a table generated from the YAML data:

Property	Description	Data Type
static_objects	List of static objects	List of objects
center	Center coordinates of the object	Tuple of [x, y]
width	Width of the object	float
height	Height of the object	float
rotation	Rotation of the object	float

In this table, I've represented the YAML data as a table with property names, descriptions, and data types for each field.

Configuring a Wall

Property	Description	Data Type
endpoints	List of endpoint coordinates	List of Tuples of XY coordinates of vertices

Note

It is highly recommended to create a starting configuration using the GUI and edit the specifics from the YAML file.