Social perception topics#

PAL robots implement the REP-155 aka ROS4HRI to represents the humans detected by the robot.

See Social perception with ROS4HRI for an introduction to this API and 👥 Social perception for details regarding our implementation.

Note

PAL OS 25.01 does not covers the full ROS4HRI specification. For instance, groups are currently not automatically recognised, and thus, the /humans/groups/ topics do not exist.

Face-related topics#

Data on faces is provided by the hri_face_detect node.

Name	Description
/humans/faces/*/roi	Region of the face in the source image
/humans/faces/*/cropped	Cropped face image, if necessary scaled, centered and 0-padded to match the `/humans/faces/width` and `/humans/faces/height` ROS parameters
/humans/faces/*/aligned	Aligned (eg, the two eyes are horizontally aligned) version of the cropped face, with same resolution as `/cropped`
/humans/faces/*/landmarks	2D facial landmarks extracted from the face

Body-related topics#

Data on bodies is provided by the hri_body_detect node, that supports up to 5 bodies (recommended), or the hri_fullbody node, that supports only one body.

Name	Description
/humans/bodies/*/roi	Region of the whole body body in the source image
/humans/bodies/*/cropped	Cropped body image
/humans/bodies/*/skeleton2d	The 2D points of the the detected skeleton
/humans/bodies/*/joint_states	The joint state of the human body, following the kinematic model of the human
/humans/bodies/*/position	Body position, representing the point between the hips of the tracked body
/humans/bodies/*/velocity	Velocity and direction of movement of the detect body

Voice-related topics#

Caution

As of PAL OS 25.01, PAL robots can only perform speech recognition on a single voice, and does not feature automatic speaker identification.

As such, the robot only ever exposes one single voice with id anonymous_speaker.

Data on voices is provided by the asr_vosk node, which performs the speech recognition.

Name	Description
/humans/voices/*/audio	Separated audio stream for this voice
/humans/voices/*/is_speaking	Whether or not speech is recognised from this voice
/humans/voices/*/speech	The live stream of speech recognized via an ASR engine

Person-related topics#

Data on persons is primarily provided by the hri_person_manager node (for the probabilistic features fusion), and hri_engagement for the engagement status.

Name	Description
/humans/persons/*/face_id	Face matched to that person (if any)
/humans/persons/*/body_id	Body matched to that person (if any)
/humans/persons/*/voice_id	Voice matched to that person (if any)
/humans/persons/*/alias	If this person has been merged with another, this topic contains the person ID of the new person
/humans/persons/*/anonymous	If true, the person is anonymous, ie has not yet been identified, and has not been issued a permanent ID
/humans/persons/*/engagement_status	Engagement status of the person with the robot
/humans/persons/*/location_confidence	Location confidence; 1 means person currently seen, 0 means person location unknown. See Person Frame