Skip to content

existing_infrastructure_from_pal_robotics.md

Latest commit

 

History

History
32 lines (28 loc) · 1.67 KB

existing_infrastructure_from_pal_robotics.md

File metadata and controls

32 lines (28 loc) · 1.67 KB

Existing Infrastructure from PAL

Hardware

  • Sensors & Perception:
  • Head Camera: RGB-D
  • Torso Front Camera: RGB-D
  • Torso Back Camera: Stereo-fisheye
  • LIDAR
  • Interaction & Display:
  • Touchscreen
  • Eyes: 2 LCD screens with custom animations
  • Ears: 2x 16 RGB LEDs
  • Back: 40 RGB LED ring
  • Audio:
  • Speakers: 2 x 30W Hi-Fi full-range
  • Microphones: 4x high-performance digital microphone array

Software

  • ROS4HRI Framework: Implements REP-155 for human representation using unique identifiers: face, body, voice, and person.
  • Face Detection and Recognition: Detects and recognizes faces, extracting facial landmarks for identification.
  • Skeleton Tracking: Supports both 2D and 3D skeleton tracking to interpret human poses and movements. Only for single body.
  • Speech Recognition: Utilizes the Vosk recognizer for offline speech-to-text conversion.
  • Dialogue Management: Integrates with the RASA chatbot engine to manage conversations.
  • Speech Synthesis: Uses the Acapela engine for text-to-speech conversion, only for English.
  • Knowledge Base: Supports RDF/OWL standards for symbolic reasoning and complex queries using SPARQL.
  • Gesture Recording and Playback: Allows for the creation and execution of predefined gestures.
  • expressive_eyes Node: Controls the robot's eye movements and expressions.
  • gaze_manager Node: Coordinates head and eye movements for natural gaze behavior.
  • Web-Based User Interface: Allows intuitive control and customization of ARI's features.
  • Simulation Tools: Provides a ROS-based simulator for testing and development without the physical robot.