Recognizing the Waving Gesture in the Interaction with a Social Robot

A Survey on Recent Advances in Social Robotics

“…In [14], a robot used deep neural networks to learn human behavior based on data it gathered during its interactions. The purpose was to use the most appropriate action among waving, looking toward human, waving, and handshaking Additionally, in the context of social intelligence, a vision-based framework for allowing robots to recognize and respond to hand waving gestures were presented in [16], increasing its social believability. Furthermore, a humanoid robot was endowed with human-like welcoming behaviors with enthusiasm for the purpose of drawing the attention of persons entering a building in [17].…”

“…It can be easily stated that the Softbank Robotics [117] Nao robot [118] has had a great share of the previous work and studies [12,25,27,29,30,36,38,39,41,48,50,54,119,120] for the different aspects of its usability as it will be shown later. Another humanoid robot of Softbank Robotics, Pepper [121], was also widely used [16,[19][20][21]43]. Other robotic platforms that can be mentioned are ASIMO [44,84,122,123], iCub [124], TERESA [15,22], Double [23,47], Furhat [26,125], Stevie [62,126], LUCA [33] which was inspired by the OPSORO platform [127,128], The Universal Robot Trainer (URT) [70], Kaspar [37,77], Mio Amico [42], Reeti [40,129], BabeBay [81], Kiddo [49], Paro...…”

“…Also, GloVe is a reliable word embedding model used in conversational systems [149,150]. Another example is Handtrack, an end-to-end trainable CNN model that learns to track the human hand [16,151]. Also, studies have been made in the aim of modeling and digitizing human emotions, which can have projections on social robot behavior and intelligence perception [152,153].…”

“…Furthermore, social robots can convey information to humans via different means such as speech, gestures, and facial expressions [11]. Advances in machine learning and pattern recognition are allowing robots to acquire realistic social behaviors involving behaviors in speech and gestures [12][13][14][15][16][17]. This enabled robots to be active conversation parties either in text-based or voice-based conversations.…”