Human-Robot Interaction
Spring 2009
School of Design, Hong Kong Polytechnic University
Team: AJ Libunao, David Lu, Michelle Li
Role: Theoretical research, concept development, LEGO construction, programming logic, testing, documentation, presentation
Emotabot is a sociable robot that interacts with a human based on the biometric feedback from the human being. Emotabot engages in interaction when it detects a person in close proximity by raising its head and giving a verbal greeting. It then reads the heart rate variability of the user via an emWave stress reliever device to infer the user’s emotional state and then responds accordingly to create the illusion of understanding using visual facial cues and speech. When the user is in a calm, mental state, that is, she has a coherent and stable heart rate, the robot will smile, open its ears, and provide positive comments. On the other hand, when the user is in a stressed state, that is, she exhibits an inconsistent heart rate, the robot will display a sad expression and tell jokes to encourage the user to relax. Conversation is facilitated between robot and human through turn-taking tactics. When Emotabot senses that the user is leaving, it says ‘goodbye’ and drops its head.Through these interactions, we believe that using a sociable robot in this particular interaction will help people better manage their physical and emotional states than with a specific monitoring device.
Emotabot was built using LEGO Mindstorm NXT for the physical form and actuating motors and uses emWave as an interface between the robot and human. The emWave has two biofeedback sensors that measure users’ heart rate variability (heart rhythm). A figure sensor is built into the emWave device that only requires users to place the index finger on it and an external ear sensor that requires users to clip it on the ears. Different LED colors represent different states of the users. We used this LED colour display to link up with the LEGO Mindstorm NXT through the Mindstorm light sensor. By having the light sensor detect colour output from the emWave, we can determine what state the user is in (green and blue corresponds to a calm state and “in-between” state, respectively, while red denotes a stressed state). An ultrasonic sensor detects the user’s proximity in relation to the robot. In this case, if the user is within 24 inches of Emotabot, the robot will raise its head to greet the person. Once the person leaves the field of view of the ultrasonic sensor, the robot will lower its head and bid the user farewell.
The heart plays an important role in affecting the emotional and cognitive state of human beings through physiological changes in the body. Therefore, theories that we draw upon for our robot creation are from the spheres of psychology, child development and social psychology. The HCI field of affective computing by which systems detect user’s physiological states is still relatively new. Many research works deal only with the user’s physiology and responding with environmental changes, for example, but not necessarily focusing responding to or changing the person’s emotions. By creating a robot that can develop a relationship based on emotional communication, verbal or non-verbal, a meaningful interaction can be achieved. With the limited resources and knowledge about biometric feedback and robotics, we can only build a very simple experimental project using the biometric feedback concept. This concept is quite interesting because it is very difficult to lie through our physiological and emotional states. By illuminating the body as the interface in this interaction, human and robot can have a more direct and valid interaction.
