irsim.msg.messages

ROS-style, dependency-free messages for simulation snapshots. Used for exchanging data with ROS and other simulators, and for logging and playback.

The classes in this module are plain dataclasses. They intentionally do not depend on ROS, which keeps them usable in regular Python, learning, and batch simulation workflows while providing familiar topic names and message shapes.

Classes

Message	Base class that provides JSON-compatible message serialization.
Header	Version-neutral metadata shared by all IR-SIM messages.
Point	ROS geometry_msgs/Point-shaped Cartesian position.
Vector3	ROS geometry_msgs/Vector3-shaped Cartesian vector.
Quaternion	ROS geometry_msgs/Quaternion-shaped orientation.
Pose	ROS geometry_msgs/Pose-shaped position and orientation.
PoseWithCovariance	ROS geometry_msgs/PoseWithCovariance-shaped pose.
Twist	ROS geometry_msgs/Twist-shaped linear and angular velocity.
TwistWithCovariance	ROS geometry_msgs/TwistWithCovariance-shaped velocity.
Odometry	ROS nav_msgs/Odometry-shaped object pose and velocity.
LaserScan	Shared sensor_msgs/LaserScan fields for a LiDAR snapshot.
ContactState	One contact of an object in the last step (collision_mode: contact).
ObjectState	Topic-shaped messages and simulator metadata for one object.
WorldState	Complete point-in-time state of an IR-SIM environment.

Module Contents

class irsim.msg.messages.Message

Base class that provides JSON-compatible message serialization.

to_dict() → dict[str, Any]

Return the complete message as JSON-compatible built-in values.

class irsim.msg.messages.Header

Bases: Message

Version-neutral metadata shared by all IR-SIM messages.

ros_type: ClassVar[str] = 'std_msgs/Header'

seq: int = 0

stamp: float = 0.0

frame_id: str = 'world'

class irsim.msg.messages.Point

Bases: Message

ROS geometry_msgs/Point-shaped Cartesian position.

ros_type: ClassVar[str] = 'geometry_msgs/Point'

x: float = 0.0

y: float = 0.0

z: float = 0.0

class irsim.msg.messages.Vector3

Bases: Message

ROS geometry_msgs/Vector3-shaped Cartesian vector.

ros_type: ClassVar[str] = 'geometry_msgs/Vector3'

x: float = 0.0

y: float = 0.0

z: float = 0.0

class irsim.msg.messages.Quaternion

Bases: Message

ROS geometry_msgs/Quaternion-shaped orientation.

ros_type: ClassVar[str] = 'geometry_msgs/Quaternion'

x: float = 0.0

y: float = 0.0

z: float = 0.0

w: float = 1.0

classmethod from_yaw(yaw: float) → Quaternion

Create a planar quaternion from a yaw angle in radians.

to_yaw() → float

Return the planar yaw angle in radians, from a normalized copy.

抛出:

ValueError -- If the quaternion is zero and defines no rotation.

class irsim.msg.messages.Pose

Bases: Message

ROS geometry_msgs/Pose-shaped position and orientation.

ros_type: ClassVar[str] = 'geometry_msgs/Pose'

position: Point

orientation: Quaternion

class irsim.msg.messages.PoseWithCovariance

Bases: Message

ROS geometry_msgs/PoseWithCovariance-shaped pose.

ros_type: ClassVar[str] = 'geometry_msgs/PoseWithCovariance'

pose: Pose

covariance: numpy.ndarray

class irsim.msg.messages.Twist

Bases: Message

ROS geometry_msgs/Twist-shaped linear and angular velocity.

ros_type: ClassVar[str] = 'geometry_msgs/Twist'

linear: Vector3

angular: Vector3

class irsim.msg.messages.TwistWithCovariance

Bases: Message

ROS geometry_msgs/TwistWithCovariance-shaped velocity.

ros_type: ClassVar[str] = 'geometry_msgs/TwistWithCovariance'

twist: Twist

covariance: numpy.ndarray

class irsim.msg.messages.Odometry

Bases: Message

ROS nav_msgs/Odometry-shaped object pose and velocity.

ros_type: ClassVar[str] = 'nav_msgs/Odometry'

header: Header

child_frame_id: str = ''

pose: PoseWithCovariance

twist: TwistWithCovariance

classmethod from_object(obj: Any, *, stamp: float = 0.0, seq: int = 0, frame_id: str = 'world') → Odometry

Capture an object's planar pose and body-frame velocity.

classmethod from_msg(msg: Any) → Odometry

Adopt any ROS-compatible nav_msgs/Odometry-shaped message.

Only the planar fields IR-SIM uses are read, so native ROS messages and partial stand-ins are both accepted. Header metadata is left out, since its layout is ROS-version specific and belongs to the receiver.

参数:

msg -- An object exposing pose.pose and twist.twist.

返回:

An independent, validated IR-SIM odometry message.

返回类型:

Odometry

抛出:

• TypeError -- If msg does not expose ROS odometry pose and twist.

• ValueError -- If any planar value is not finite.

to_state_velocity(obj: Any) → tuple[numpy.ndarray, numpy.ndarray]

Convert this odometry into state and velocity arrays for obj.

The inverse of from_object(): the planar pose replaces the first three state values, and the body-frame twist is mapped onto the velocity layout of the object's kinematics. obj is only read, so a caller can validate several updates before applying any of them.

Ackermann steering is carried as a yaw rate, which is zero whenever the car is not moving. A stopped car therefore conveys no steering angle, and obj keeps the one it already had.

参数:

obj -- The simulation object whose layout the arrays must match.

返回:

The new (state, velocity) arrays for obj.

返回类型:

tuple

抛出:

ValueError -- If obj has fewer than three state values, an Ackermann object has no steering value, or the orientation defines no rotation.

class irsim.msg.messages.LaserScan

Bases: Message

Shared sensor_msgs/LaserScan fields for a LiDAR snapshot.

ros_type: ClassVar[str] = 'sensor_msgs/LaserScan'

header: Header

angle_min: float

angle_max: float

angle_increment: float

time_increment: float

scan_time: float

range_min: float

range_max: float

ranges: numpy.ndarray

intensities: numpy.ndarray

classmethod from_sensor(sensor: Any, *, stamp: float = 0.0, seq: int = 0, frame_id: str | None = None, use_inf: bool = False) → LaserScan

Capture a LiDAR sensor without sharing its mutable arrays.

By default ranges are the sensor's own finite values, range_max for a miss, so snapshots used for learning are unchanged. With use_inf=True a beam without a usable return becomes +inf when nothing came back within range_max and -inf when the return was too close to measure, as Gazebo publishes it under REP 117, so ROS consumers read the scan as they would a real driver's.

class irsim.msg.messages.ContactState

Bases: Message

One contact of an object in the last step (collision_mode: contact).

normal points from the partner toward the object, point is the world-frame contact point, depth the overlap that was pushed apart (zero or negative for a resting contact) and force the contact force in newtons along the normal.

ros_type: ClassVar[str] = 'irsim_msgs/ContactState'

other_id: int = -1

other: str = ''

point: Point

normal: Vector3

depth: float = 0.0

force: float = 0.0

classmethod from_report(report: Any) → ContactState

Serialize an object's ContactReport.

class irsim.msg.messages.ObjectState

Bases: Message

Topic-shaped messages and simulator metadata for one object.

In collision_mode: contact the state also carries what a contact sensor reports: the net contact_force, one ContactState per contact, and how long the object has been touching (contact_time) or free (air_time).

header: Header

id: int

name: str

role: str

kinematics: str | None

shape: str

odom: Odometry

scan: LaserScan | None

goal: numpy.ndarray | None

static: bool

arrive: bool

collision: bool

collision_ids: list[int] = []

scans: list[LaserScan] = []

contact_force: Vector3

contact_time: float = 0.0

air_time: float = 0.0

contacts: list[ContactState] = []

classmethod from_object(obj: Any, *, stamp: float = 0.0, seq: int = 0, frame_id: str = 'world', use_inf: bool = False) → ObjectState

Capture an object as conventional odom and scan topics.

use_inf is passed to LaserScan.from_sensor().

property sensors: list[LaserScan]

Return all LiDAR messages as a compatibility alias for scans.

class irsim.msg.messages.WorldState

Bases: Message

Complete point-in-time state of an IR-SIM environment.

header: Header

name: str

status: str

step_time: float

objects: list[ObjectState] = []

classmethod from_env(env: Any, frame_id: str = 'world', use_inf: bool = False) → WorldState

Capture the current state and sensor data from an environment.

use_inf is passed to every LaserScan.from_sensor().

property robots: list[ObjectState]

Return the robot messages in this snapshot.

property obstacles: list[ObjectState]

Return the obstacle messages in this snapshot.