The 3rd International Conference on Ubiquitous Robots and Ambient Intelligence (URAI 2006)
ٻ
An Embedded Localization Sensor Based on IR Landmark for Indoor Mobile Robot ٻ
Byungkon Sohn, Jaeyeong Lee, Heesung Chae, and Wonpil Yu Intelligent Robot Research Division, Electronics and Telecommunications Research Institute {bksohn, jylee, hschae, ywp}@etri.re.kr ٻ ٻ Abstract ڈThe localization for mobile robot in indoor environment is one of the most important issues in robot research area. This paper describes an embedded localization image sensor which computes its location by using the pixel positions of infrared sources on IR landmark. The proposed localization sensor can operate robustly on the change of illumination in common indoor environment. The components of embedded localization sensor and its performance will be mainly depicted. ٻ Keywords ڈlocalization, navigation, artificial landmark, mobile robot.
(a) (b) Fig. 2. (a) Localization sensor design, (b) Landmark design
1. Introduction
ٻ ٻ ٻTo make the robot navigate by itself and plan paths on its way to a goal, the robot should have information about its location. The localization technique in mobile robot is to find out the 2-D position and heading angle of robot in a certain space that is built for robot localization. Localization technology in the field of mobile robotics has been well studied and a multitude of methods have been proposed so far; a good overview on the robot localization technology can be found in [2]. Sensor for the robot localization has the following properties: accuracy, repeatability, real-time computation, scalability, robustness, and economic efficiency. Our localization sensor suite meets those needs above.
2. Architecture of Sensor Suite Localization sensor (Fig. 2. (a)) is composed of micro processor, image sensor, RF (radio frequency) communication chip, and SDRAM memory. Micro processor includes embedded software to control the image sensor, RF chip, UART and SDRAM memory. It also has the algorithm detecting IR landmarks: the sensor computes its location in real-time by tracking the IR source. Sensor RF Chip
Micro Processor
Serial
Wireless RF communication
IR Landmark RF Chip
Micro Processor IR LED
Image Sensor Serial communication
PC
Fig 1. Architecture of embedded localization sensor
Image sensor has its own registers to adjust image resolution, image color depth, and camera exposure. We used monochrome image with 240x240 resolution. The RF chip (2.4GHz) controls the wireless IR landmarks. The sensor can not only turn the landmarks on and off with RF communication, but also the sensor can receive the information about the remaining battery power of the landmarks. The SDRAM memory stores the data that is used during computing location. The sensor computes its location and it transfers its location data to robot via UART. The sensor has dimension of 37x37x42.6(mm) and the proper size enables the sensor to be installed on any type of robot. The IR landmark (Fig. 2. (b)) is composed of microprocessor, RF chip, and IR LED. Landmark uses two AAA battery as its power. The microprocessor on landmark controls RF communication and IR LED.
3. Localization The localization sensor is configured such that infrared landmark modules are attached on the ceiling of a space and the sensor is mounted on top of a mobile robot as shown in Fig. 3. The CMOS camera in the sensor detects IR landmark through an infrared bandpass filter. It is oriented to look upward so that its optical axis is perpendicular to the ground. In order to obtain maximal field of view, a wide-angle camera lens is utilized. The location information of the sensor can be obtained when at least two IR LEDs are detected within the field of view of the camera. The localization is performed in two steps. In the first step, the image coordinates of the IR LED are computed and landmark IDs are identified by the sensor. In the
452
The 3rd International Conference on Ubiquitous Robots and Ambient Intelligence (URAI 2006)
ٻ with standard deviation of 2.9cm. We have the maximum position error of 17.1 cm at (180, 60) because of uneven ground condition. The experimental result shows that the sensor suite gives acceptable localization performance in indoor environment. The field of view of the sensor when the ceiling height is 2.5m is 5x5(m). Table 1 shows the result of the repeatability (10 times) of localization when a robot stayed at the same position of (0,0) and (200, 200). On the heading angle experiment, we turn the sensor on the same position by 10 degree from 0 degree to 350 degree. The angular mean error is 0.21 degree and its deviation is 0.31. From the above experiment results, we can conclude that our embedded sensor is very effective as a robot localization sensor.
Landmark j
Landmark i
Ceiling
d ij
h Image plane ( xi , yi ) ( xi , y j )
f Floor Robot (Camera)
Fig. 3. Sensor configuration of the localization system second step 2-D location and heading angle of the sensor are computed from the relationship between the image coordinate and world coordinate of the detected LEDs. The location update rate of the sensor is 30Hz. The details about the localization algorithm can be found in [7].
5. Conclusion In this paper, we have proposed a localization sensor for mobile robots. Important factors for localization sensor are accuracy, scalability, robustness, and economic efficiency. Experimental results show that the proposed sensor satisfies the factors mentioned above.
4. Experimental Results Acknowledgements We have done the experiments to focus on the accuracy and repeatability of the estimated position with the sensor discussed above. The height of the ceiling is 2.6m and each distance between infrared landmark is 1.20m. The location information was computed for every grid points which are spaced regularly with 60cm distance.
Fig. 4. Static localization results of the fixed points Fig. 4 shows the localization result using the sensor in the fixed points when the heading angle is zero. The estimated positions are represented by asterisk and true positions with circle. The mean position error is 4.1cm Table 1 Repeatability result Position Mean Deviation (0,0) (-1.34,1.997) 0.0317 (-200, 200) (-196.47,203.42) 0.1289
This work was supported by Ministry of Information and Communication, Korea.
References [1] I. J. Cox and G. T. Wilfong, Autonomous Robot Vehicles, Springer-Verlag, 1990. [2] J. Borenstein, H. R. Everett, and L. Feng, “Where am I? Sensors andMethods for Mobile Robot Positioning,” Technical Report, Univ. of Michigan, APR. 1996. [3] Fuji-Keizai USA, Inc., “Wireless Sensing Networks: Market, R&D and Commercialization Activities,” Market Research Report, FEB. 2004. [4] K. Pahlavan, X. Li, and J. Makela, “Indoor Geolocation Science and Technology,” IEEE Communications Magazine, pp. 112-118, FEB. 2002. [5] T. S. Rappaport, J. H. Reed, and B. D. Woerner, “Position Location Using Wireless Communications on Highways of the Future,” IEEE Communications Magazine, pp. 33-41, OCT. 1996. [6] K. Yamano, K et al., “Self-localization of mobile robots with RFID system by using support vector machine,” IEEE Int. Cont. Intell. Robots and Systems, pp. 3756-3761, 2004. [7] J. Y. Lee, H. S. Chae, and W. P. Yu, “A Real-time Localization System Based on IR Landmark for Mobile Robot in Indoor Environment”, Journal of Contro, Automation and Systems Engineering Vol. 12, No.9, p.868-875, 2006 ٻ
453