How Does It Work?
aRadar® utilizes 24 GHz wireless signals reflected off people’s bodies to measure heartbeats and respiration with high accuracy comparable to that of an ECG monitor. The EPU III machine learning engine studies the waveforms within each heartbeat to match a person’s behavior to how they emotionally responded previously.
Emotion-detection methods available at present are based on audio-visual cues, tone of the voice, semantic analysis or on-body sensors. Though all these techniques have their advantages, they also come with limitations in terms of accuracy. We all understand how facial expressions can be deceptive and related to a certain culture or acting performance. As an example, a smile does not necessarily reflect a person’s happiness. On-body sensors such as chest bands and ECG monitors are invasive solutions. aRadar users will be relieved from the inconvenience of using any patches or wires to connect to the device.
aRadar was designed to balance the downsides of the existing emotion-detection techniques, based on the fact that human beings can lie or misrepresent facts in their languages (body, semantic). It sends wireless signals that reflect off a person’s body and come back to the device. Its beat-extraction and respiration detection algorithms break the reflections into individual spatio-temporal (x,y,z,t) heartbeats and analyze the small variations in heartbeat intervals to determine their levels of arousal and positive affect. The measurement of these almost imperceptible heart beat variations is critical to aRadar’s capability to detect emotion. A person whose signals correlate to low arousal and negative affect is more likely to be tagged as sad and vice versa.
aRadar® & EPU III
A comparative study of aRadar with vision-based “Emotion API” that focuses on facial expressions has revealed that aRadar is significantly more accurate in detecting joy, sadness, and anger.
The aRadar system’s functionality relies on a radar technique known as the Frequency Modulated Carrier Waves, which is particularly powerful because of its ability to eliminate the reflections from static objects and other humans. This high-precision body tracking is sensitive enough to pick up the rising and falling of the chest during breathing as well as the minute vibrations caused by blood pulsing through the body. As heart contractions happen much faster than breathing acceleration, measurements are used to isolate the fainter heartbeat signals.
The aRadar is specially designed to work in combination with Emoshape EPU III (Emotion Processing Unit) by feeding the EPU with raw data generated by the aRadar. Therefore, Emoshape Machine learning technology inside the emotion chip combining many stimulus can achieve a much higher accuracy in the synthetisys of the twelve primary emotions for intelligent machines: anger, fear, sadness, disgust, indifference, regret, surprise, anticipation, trust, confidence, desire and joy.
aRadar is the fundamental part that allows the emotion chip the most accurate human understanding. Incorporating four real-time inputs together with viz. face, semantic, harmonic of the voice and aRadar the EPU III through wave computing can work in harmony to create a more accurate appraisal to deliver a meaningful emotional response in the form of emotion synthesis. We have no doubt that aRadar can make a significant positive impact in numerous fields, including affective computing, AI, robotics, IoT, self driving cars, security, defense , advertising and healthcare.