Video surveillance patrol robot system in 3G, Internet and sensor networks

Weijia Jia, Fung Po Tso, Lizhuo Zhang, Weijia Jia 0001Sensys 2007 · DOI

TL;DR

This paper introduces a novel system for ubiquitous video surveillance using a 3G-enabled patrol robot integrated with wireless sensor networks and the internet. The system allows remote control and real-time video streaming to users via 3G phones or PCs, enabling proactive event detection and response.

What Problem Does This Paper Address?

The authors tackle the challenge of creating a flexible and pervasive surveillance system that can operate in diverse environments and respond to events in real-time. Traditional surveillance systems often lack mobility, remote accessibility, or the ability to proactively detect and investigate anomalies. This paper aims to bridge this gap by combining the strengths of mobile robots, wireless sensor networks for event detection, and modern communication technologies like 3G and the internet for seamless remote operation and data access. The goal is to provide a system that can patrol an area, detect unusual events through sensors, and then dispatch a robot to investigate and stream live video, all controlled remotely.

Methods

The core of the system is a central gateway that acts as an intermediary, interconnecting three distinct networks: a wireless sensor network, a 3G network, and the TCP/IP internet.

  • Sensor Network: Distributed sensors (e.g., heat, voice, light) detect abnormal events and report their readings to a sink node, which then forwards them to the central gateway.
  • 3G Patrol Robot: This mobile robot is equipped with a 3G terminal (a 3G phone). This phone serves a dual purpose:
  • Receiving control signals from remote users (via 3G phone or PC) to execute movement commands.
  • Transmitting real-time multimedia (video) streams back to the user over the 3G network or to the gateway for internet relay.
  • Remote Control & Monitoring:
    • 3G Phone: Users can receive SMS alerts, control the robot, and interact with the real-time media stream.
  • Laptop/PC: Users can query sensor data, control the robot, and receive the video stream via the internet.
  • Gateway Functionality: The gateway is crucial for orchestrating the system. It receives sensor data, analyzes it, sends SMS alerts upon event detection, accepts sensor data queries from the internet, and facilitates communication between internet users and the 3G robot.

Datasets

Not discussed in the paper. The system is demonstrated with a pilot deployment in a room with four sensors.

Key Insights and Ideas

  • Unified Network Architecture: The paper highlights the power of a central gateway in seamlessly integrating disparate networks (sensor, 3G, internet) for a cohesive surveillance solution.
  • Proactive Surveillance: Moving beyond passive monitoring, the system enables proactive investigation by dispatching a robot to an event location based on sensor alerts.
  • Remote Accessibility: The system offers flexible remote access via both 3G phones and internet-connected PCs, catering to different user preferences and network availability.
  • Interoperability: The design emphasizes interoperability, allowing for heterogeneous communication between various network types and devices.

Key Results

The paper presents a functional demo scenario illustrating the system's capabilities. In this scenario:

  • A sudden change in light intensity is detected by a sensor.
  • The central gateway receives this data and sends an SMS alert to the user, including a phone number to call for real-time video.
  • The user can then dial the robot using a 3G phone or connect via an internetworking laptop.
  • Upon connection, the user can control the robot's movement and view the real-time video feed from the robot's perspective.

The paper does not present quantitative performance metrics but rather focuses on the successful demonstration of the system's architecture and functionality.

Limitations and Future Work

Limitations:

  • The paper acknowledges that direct dialing to the robot via a 3G phone is dependent on the user's current location having 3G service coverage. If coverage is absent, internet access is required.
  • The system relies on the availability of 3G network coverage for the robot's operation and remote communication.

Future Work:

The authors suggest that the architecture supports a number of extended applications, including:

  • Remote Diagnosis: Monitoring patient or elder health status with sensors, triggering alerts to hospitals/doctors, and allowing remote diagnosis and threshold adjustments.
  • Real-time Video Surveillance: Using camera sensors in homes/offices to detect human motion, stream video to a base station, and then relay it to a 3G user for live viewing and interaction.

Summary updated

BibTeX
@inproceedings{jia2007video,
  author = {Weijia Jia and Fung Po Tso and Lizhuo Zhang and Weijia Jia},
  title = {{Video surveillance patrol robot system in 3G, Internet and sensor networks}},
  booktitle = {Sensys},
  year = {2007},
  doi = {10.1145/1322263.1322312},
}
Figure