I. Introduction

Seventeen years after the first system appeared in 1994 [3], Internet-based telerobot has made a great contribution to the modern life allowing us to remotely visit museums, tend gardens, navigate undersea, float in blimps, or handle protein crystals[1]–[5]. Whereas early researches tried to answer the question of how to control a robot through the Internet [3][6][7], recent studies have focused on how to control it in real time and deal with the inevitable Internet transmission delay, delay jitter and non-guaranteed bandwidth, etc [8]-[10]. To be effective, a research project may focus on only one single specific aspect and a common experimental platform is usually desirable for the purpose of implemental verification; however, significant work will be needed to build an experimental platform from scratch. Several Internet-based robot platforms therefore have been proposed with their strengths and limitations [9][11][12].

In [11], a web-based telerobot framework is developed in which communications between users and the robot are centered around a web server. The system consists of four modules: a commercial Pioneer mobile robot, a visual feedback display, a global environment map and a web interface, communicated over TCP protocol. By using the web interface, users are able to control the robot over the Internet to explore a laboratory or to push a ball into a goal. The use of TCP which was originally designed for the reliable transmission of static data such as e-mails and files over low-bandwidth, high-error-rate networks as the communication protocol, however, may limit the system from future developments in which the real-time attribute is highly demanded. In addition, the lack of autonomous mechanisms may influence the robot safety and downgrade the system performance in case of network congestion or interruption.

A more flexible and extensible approach is to use client-server architecture as described in [12]. This modular structure allows users to quickly construct further developments of Internet mobile robot. The system uses UDP as the transport protocol and includes essential modules for an Internet robot system such as a mobile robot, a visual feedback display, a virtual environment display and a user-friendly graphic user interface.

On a similar note, Dawei et al proposed a quite complete online robot platform in which an Omni-directional mobile robot with a five DOFs arm is controlled over the Internet by using a virtual represent [9]. During the control session, virtual environment of the remote site is continuously updated at the local site and the next position of the robot is predicted and pointed out in the virtual environment. This combined with a visual feedback display enables user to effortlessly navigate the mobile robot in an unknown environment. In addition, the robot safety is strictly ensured by build-in autonomous mechanisms. The use of sonar sensors with a measuring range of 4cm to 400cm for building the virtual environment, however, may limit the applicability of the system to indoor applications only.

In this paper, we propose a novel platform for Internet-based mobile robot systems with improvements in hardware configuration and software development. The system is in Client-Server mode, which contains users, as the command input and the Multi-Sensor Smart Robot (MSSR), as the controlled plant. The MSSR has accurate motion control with PID algorithm and contains many types of sensors to support diverse purposes of development. The MSSR is connected to the Internet via 3G mobile network. Autonomous mechanisms including obstacle avoidance and safe-point achievement are implemented in the robot. The software has two main modules: a client controller at the user site and a server module at the robot site. A multi-protocol model is applied to deliver the data exchanged between the client and server. The platform is implemented on the real Internet and the experimental result is promising. By adopting this, it will be very easy to construct an experimental system for the research on diverse teleoperation topics such as remote control algorithms, interface designs, network protocols and applications etc.

The paper is organized as follows. Details of the hardware configuration is described in Section II. The software development is described in Section III. Section IV introduces experiments in the real Internet environment. The paper concludes with an evaluation of the system, with respect to its strengths and weaknesses, and with suggestions of possible future developments.

II. Hardware Design 

In order to be a platform for different developments, the system hardware needs to be designed to support not only a specific task but also a wide range of applications including both indoor and outdoor environments. In our system, the hardware design is split into three perspectives: the communication configuration, the sensor and actuator, and the user interaction; each is developed with the feasibility, the flexibility and the extendibility in mind. Fig.1 shows an overview of the system.

A. Communication configuratuon

Figure 1.  System hardware configuration

To the best of our knowledge, most current Internet robot systems use a common configuration for the network connection in which the robot and components are connected to the Internet through a central wireless access point (fig.2) [9][11][12]. This configuration is easy to set up but it restricts the operational area of robot and components to a radius of several hundred meters due to the transmit power limitation of the wireless access point. This range is acceptable for indoor environment but is insufficient for outdoor applications such as traffic control and disaster rescue. 

Figure 2.  Network configuration of current Internet-based robot systems

In our system, instead of using a wireless access point, the 3G mobile network is utilized as the communicating bridge between the robot and the Internet. A 3G USB device with an internal mobile SIM card is used. The USB is plugged into the computer inside the robot and is registered to a mobile phone service provider allowing it to have access to the Internet (fig.1). This simple configuration enables the robot to connect to the Internet without any restrictions in physical distance as far as the 3G mobile signal is presented which is almost everywhere in the country due to the fact that the 3G signal already covers it all.

B. Sensors and Actuators

The sensors and actuators are included in a Multi-Sensor Smart Robot (MSSR) developed by our laboratory. The scheme in fig.3 describes sensors, actuators and communication channels in the MSSR. It contains basic components for motion control, sensing and navigation. These components are drive motors for moving control, sonar ranging sensors for obstacle avoidance, compass and GPS sensors for heading and global positioning, and laser range finder (LMS) and visual sensor (camera) for mapping and navigating. 

Figure 3.  Sensors, actuators and communications in the MSSR

The drive system uses high-speed, high-torque, reversible-DC motors. Each motor is attached a quadrature optical shaft encoder that provides 500 ticks per revolution for precise positioning and speed sensing. The motor control is implemented in a microprocessor-based electronic circuit with an embedded firmware which permits to control the motor by a PID algorithm.

The positioning and heading modules contain a CMPS03 compass sensor and a HOLUX GPS UB-93 module [14][15]. The compass sensor has the good heading resolution of 0.1°. The GPS with lower accuracy is used for positioning in outdoor navigation. Due to the networking is available in our system, an Assisted GPS (A-GPS) can be also used in order to locate and utilize satellite information from the network in poor signal conditions.

The MSSR provides eight SFR-05 ultrasonic sensors split into four arrays, two on each, arranged at four sides of the robot. The measuring range is from 0.04m to 4m.

On the front side of the robot, a 3D-image capturing system is built based on a SICK-LMS 221 2D laser range finder [13][22]. The system has the horizontal view angle of 100° (angle resolutions are 0.25°, 0.5° and 1°) and the vertical (pitching) view angle of 25°. The data produced by ultrasonic sensors and laser range finder which covers a range from 0.04m to 80m is used to build global and local maps of the robot’s operational environment.

The visual system is detachable and mounted on the head of the MSSR. It mainly consists of a Sony EVI-D100 pan-tilt-zoom (PTZ) color camera and an EasyCap adapter, which is to capture the video. The rotation ranges of the pan-tilt camera are from -100 to +100 degrees in horizon, from -25 to +25 degrees in vertical and are available to give the user a clear view of the environment in front of the robot.

The communication data between devices and the computer in the robot is transferred via several channels: low-rate channels with standards of RS-485 and RS-232 and high-rate channels with USB ports. Devices using the RS-485 are managed by an on-board 60MHz Microchip dsPIC30F4011-based microcontroller with independent controller boards for a versatile operating environment. A RS-485 bus is established to maintain the communication between controllers and reserve the expansibility to support various accessories. Devices using the RS-232 are directly connected the USB-to-COM modules. Commands of control and acquisition with short messages are realized in low-rate channels. On the other side, images from camera are captured by a frame grabber and directed to a high-rate USB port. The communication between the remote-site (robot) and client-site (user), as described previously, is realized by computer network.

C.User interaction devices

sonal computer and a joystick. The computer is an ASUS notebook computer with 1.5GHz M-Centrino processor, 500Mb RAM and Windows XP operating system. The computer, with installed control software, allows users to retrieve feedback information of the remote site and navigate the robot to explore an unknown environment. To support users with a more convenient way of control, a joystick is added. The joystick is the 3D Logitech Extreme series with 10 bit resolution in horizontal and vertical axes and 12 functioning buttons. In the system, the joystick interprets users’ inputs to a sequence of control parameters and forwards them to the control software for processing.

III. Conclusion

It is extremely time-consuming to build an experimental platform for the study of Internet robots from scratch. In this paper, a new modular platform for Internet mobile robotic systems is developed. The system hardware mainly consists of a Multi-Sensor Smart Robot. Many types of sensors including position speed encoders, integrated sonar ranging sensors, compass and laser finder sensors, the global positioning system (GPS) and the visual system are implemented allowing the robot to support a wide range of applications including both indoor and outdoor environment. The limitation in working area is removed by the use of 3G mobile network. The system employs a client-server software architecture for robot control and feedback information display. The exchanged data between the client and server is transmitted over the Internet by a multi-protocol model. Autonomous mechanisms based on fuzzy logic algorithms are implemented to ensure the robot safety. The platform has been tested in different environments, and the results are promising.

                                                                                                                                          T.S Trần Thuận Hoàng