"Tom" pet robot applied to urban autism

Xingqian Li, Chenwei Lou, Jian Zhao, Huapeng Wei, Hongwei Zhao

Abstract

With the fast development of network information technology, more and more people are immersed in the virtual community environment brought by the network, ignoring the social interaction in real life. The consequent urban autism problem has become more and more serious. This paper proposes a solution by promoting offline communication and eliminating loneliness through emotional communication between pet robots and breeders. We developed a design called "Tom," a smart pet robot integrated with a social mechanism named "Tom-Talker." This platform encourages users to socialize offline through a reward mechanism and a friend recommendation algorithm. Additionally, "Tom" utilizes an emotional interaction algorithm to recognize user emotions and simulate animal-like responses. Experimental results indicate that the system effectively addresses urban autism by increasing real-world social engagement.

1. Introduction

The researchers identify a growing sociological trend termed "urban autism": a state where individuals are hyper-connected through social media but lack meaningful, high-quality interpersonal relationships in real life. While traditional pets have long served as a remedy for loneliness, the barriers of maintenance, disease, and urban living spaces make pet ownership difficult for many. Existing "companion robots" often fail because they lack genuine emotional reciprocity or focus solely on entertainment. This paper seeks to bridge the gap by creating a robot that not only simulates animal-like emotional responses but also acts as a catalyst for "vertical socialization"—connecting people with shared interests (robotic pets) to meet in physical spaces.

2. System Architecture

2.1 Hardware Design

The "Tom" robot is built on a hybrid hardware architecture designed for both high-level processing and real-time sensor management. A Raspberry Pi serves as the central brain, handling computationally intensive tasks such as computer vision for emotion recognition and network communication with the social platform. Complementing this, an Arduino acts as the peripheral nervous system, interfacing with a suite of sensors including smoke, light, temperature, and ultrasonic distance sensors to monitor the environment and the user's proximity.

Hardware block diagram of the Tom robot, utilizing a dual-controller architecture with Raspberry Pi and Arduino.

2.2 The Tom-Talker Social Platform

The "Tom-Talker" platform serves as the social glue of the system. It facilitates a unique interaction model where the physical robot acts as an intermediary between the digital platform and the user's real-world social life. The platform uses an improved collaborative filtering algorithm that prioritizes geographical proximity and "social circle intersection" to recommend friends. It issues offline tasks—such as meeting another "breeder" at a local park—and utilizes a Logical Growth Model for rewards to incentivize physical movement and social expansion.

Overview of the Tom-Talker social platform architecture, facilitating interactions between users and their pet robots.

3. Emotion Recognition and Simulation

To ensure genuine emotional reciprocity, the system employs a Convolutional Neural Network (CNN) based on the Xception architecture for facial expression recognition. To optimize performance for the Raspberry Pi, the team utilized depthwise separable convolutions and batch normalization, allowing for real-time processing of seven facial expressions (anger, disgust, fear, happy, sad, surprise, and neutral) using the Kaggle FER dataset.

Beyond recognition, Tom simulates animal-like emotions using a probabilistic model and a dopamine simulation algorithm. The robot’s mood is dynamic, shifting based on environmental comfort data from its sensors and the frequency of positive interactions with the user.

4. Key Insights and Social Mechanics

  • Social Stagnation Prevention: The reward mechanism includes a "threshold" where reward growth slows down as interaction with a single individual increases. This mathematically nudges users to expand their social circles rather than remaining in isolated social silos.
  • Vertical Socialization: By focusing on the specific niche of pet ownership, the platform creates a lower barrier to entry for social interaction compared to general-purpose social networks.
  • IoT Integration: The robot functions as an IoT node, using NFC for "feeding" virtual props and Bluetooth/Wi-Fi to synchronize social tasks, making the physical object a tangible extension of the digital social experience.

5. Experimental Results

The system was validated through two primary lenses: social impact and technical performance.

  1. Social Impact: In a study of 400 participants, users of the "Tom-Talker" model saw their average weekly social time increase significantly from approximately 12 hours to 18 hours. Furthermore, the average social circle size grew from 9 to 15 individuals.
  2. Technical Performance: The emotion recognition model achieved high precision, particularly for "happy" (83%) and "neutral" (63%) states.
  3. User Satisfaction: Qualitative surveys of 20 users indicated high satisfaction with the robot's emotional responses, suggesting the dopamine-simulation model effectively mimics natural interaction.

6. Conclusion and Future Work

While "Tom" successfully demonstrates the potential for robotic companions to drive offline socialization, limitations remain. The current emotional interaction studies involve small sample sizes, and the physical design of the robot could be further refined to enhance long-term aesthetic appeal. Future research will focus on long-term user dynamics to ensure that the social habits formed are sustainable beyond the initial novelty period.

Summary updated

Abstract

With the fast development of network information technology, more and more people are immersed in the virtual community environment brought by the network, ignoring the social interaction in real life. The consequent urban autism problem has become more and more serious. Promoting offline communication between people " and "eliminating loneliness through emotional communication between pet robots and breeders" to solve this problem, and has developed a design called "Tom". "Tom" is a smart pet robot with a pet robot-based social mechanism Called "Tom-Talker". The main contribution of this paper is to propose a social mechanism called "Tom-Talker" that encourages users to socialize offline. And "Tom-Talker" also has a corresponding reward mechanism and a friend recommendation algorithm. It also proposes a pet robot named "Tom" with an emotional interaction algorithm to recognize users' emotions, simulate animal emotions and communicate emotionally with use s. This paper designs experiments and analyzes the results. The results show that our pet robots have a good effect on solving urban autism problems.
BibTeX
@misc{li2019tom,
  author = {Xingqian Li and Chenwei Lou and Jian Zhao and Huapeng Wei and Hongwei Zhao},
  title = {{"Tom" pet robot applied to urban autism}},
  year = {2019},
  eprint = {1905.05652},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1905.05652},
}
Figure