PaPeRo 2005 is the first robot with a taste! It can walk, talk and recognizes people both visually and through audio interfaces. It features a pair of CCD's for 'eyes' and has 8 separate microphones for determining the direction sound is coming from and 9 touch sensors.
PaPeRo changes its reactions to each person depending on how it is approached. In addition, it has models for action reflecting initiative, desire, etc. and it changes its character depending on its interaction with people. It will repeat actions that it has been praised for.
The robot will engage in autonomous activities such as walking around the room at will, connecting to the Internet and obtaining the latest information or dancing on its own. The autonomous activities it engages in vary according to its character.
This robotic companion can announce the weather report, read the news, communicate the predictions (overall fortune, work fortune etc.), play various games.
It is equiped with tv remote control signal transmitter (can act as an tv remote control), multiple touch sensors, multiple microphones, 2 CCD cameras, LEDs, speakers, ultrasonic, floor and lift sensors, built-in Internet wireless modem handset, 2 USB 2.0 ports, Intel Pentium M 1.6GHz, 512MB RAM, 40GB HD, 1 VGA out 2 PCMCIA Type II card slots. It basically is a PC with additional sensors. This stand-alone architecture means it can be easily taken out of the laboratory and into many different locations, environments and situations.
The maximum speed of movement is 20 cm per second.
PaPeRo was named after an acronym of "Partner-type Personal Robot."
This personal robot is equipped with a speech signal processing and recognition technology that enables the robot to precisely locate the speech source, and also to recognize and understand commands in a noisy home environment or even while the robot itself is talking. It can differenciate around 650 words and can speak approximately 3000 words.
Handwritten character recognition technology allows this robot to read handwritten characters, remember names, and calculate mathematical formulae.
Video recognition technology enables a variety of interactions such as recognition of waving hands and objects such as a moving pendulum. It can find and identify people's faces and track a face it has found.
Nine touch sensors are mounted on the head and the body of the robot to achieve various interaction.
The PaPeRo robot can link with other robots or with remote terminals (like PC) via a network.
This robot can be used in a childcare environment. It features communication capability targeted at children, speech recognition enabled in noisy environments such as nursery schools or kindergartens, simultaneous speech recognition for multiple utterances with wireless microphones. The robot can recognize what is said by who out of simultaneous multiple utterances with the help of a special jacket worn by the child, which is equipped with a wireless microphone and an ultrasound transmitter.
Enhanced recognition of children's faces enables precise identification of face direction and an enhanced database of children's faces. Precise face-direction identification is effective for locating children who can't keep still in front of a camera.
An ultrasonic child locator can locate a child with only an error of a few centimeters within the range of 5 meters by reception of the ultrasounds radiated from the special jacket worn by the child.
PaPeRo has multiple play functions for children: singing along with the robot, chatting with the robot, face registration/recognition, robot quizzes, making greetings, connection with mobile phones, robot dances, and roll calls.
Reduced motor power, no-finger-catching design, proximity sensors, and bumper switches immediately stop the robot upon detection of contact, realizing safe interaction even with a three-year-old child. This robot enables remote child monitoring and conversation through the hardware such as its eyes, mouth and ears with the help of a mobile phone network.
The Papero beams infrared rays of various wave lengths and depending on the absorbency spectrum which is unique for every food item, it determines the taste by comparing the values. It is so accurate that it can distinguish Cheese by it's brand.
The robot automatically changes its mode (character) depending on the way it connects with people. Each mode is expressed by the way the robot speaks, moves etc. The five characters are: leader (does what it likes), knowledgeable (it informs you), dancing, lazy and computer.