Remotely Controlled Video Surveillance Monitoring System Using Microsoft Robotics Studio

Abstract

Robots consist of several mechanical and electronics parts. To accomplish a task, concurrency and coordination between these parts are required. The complexity of this combination makes programming of a robot complicated and time consuming. Robotic development environments (RDEs) were introduced to address this issue by providing a framework for robotics developers, which consists of programming tools, program components, and solutions to common robotics challenges. Robotic programmers can take advantage of the tools shipped with an RDE and use already developed components in their programs. These robotic frameworks offer some solutions such as teleoperation, distribution of processes, and predictive display. Some of them also provide a simulation environment where developers can test their codes on a simulated model of a robot, without creating or purchasing the actual robot. In this thesis, some of the most popular RDEs are briefly reviewed and one is selected. Microsoft Robotics Development Studio (MRDS) is the RDE that was chosen to program an existing wire-suspended video monitoring surveillance robotic system. The service-oriented architecture of MRDS makes all programs developed in this framework capable of working with the network and using the Internet as an infrastructure for teleoperation. Also, the same architecture allows distribution of processes (services) between multiple computers. In addition, the powerful simulation environment of MRDS could be used as a predictive display, allowing operators to predict and monitor the reaction to their commands; when the simulated model of the robot responds to the operator's commands at the same time as the actual robot. The goal of this thesis was to validate several solutions of MRDS and examine its ease of use and its helpfulness in speeding up the implementation of robotics programs. Despite the capabilities these RDEs provide, they present unique challenges for programmers. This experiment was completed by developing new MRDS services, using the existing components, and customizing the existing codes. Some of the features of MRDS are demonstrated in the final project, letting the operator control the robot over the Internet simultaneously with the simulated model of the robot. Also, distribution of processes between several computers is tested successfully. The experiment showed that the selected RDE has powerful capabilities and helpful tools for developers but it has a long learning curve due to its complexity which also slows down the development of services and hence the delivery of the final product.