Autonomous Positioning of Mobile Robot Based on RFID Information Fusion Algorithm

Transcription

1 Autonomous Positioning of Mobile Robot Based on RFID Information Fusion Algorithm Hua Peng ChongQing College of Electronic Engineering ChongQing College, China Abstract To improve the mobile performance of the robot. Establishing the automatic positioning system of mobile robot based on RFID and information fusion technology. The paper described the development of mobile robot, introduced the relative theory of the robot technology, established the mathematical model of the positioning system, and constructed the mobile robot positioning platform based on RFID technology. The paper studied the commonly used positioning algorithm, constructed the corresponding experimental system, and found that the mobile robot localization system in the article had a good positioning effect through the analysis of specific experimental data. The mobile robot localization system based on RFID and information fusion technology can achieve the design goal and has broad application prospect. Keywords - pore pressure; fire; self-consolidating concrete; fiber reinforcement; spalling I. INTRODUCTION Application of multi sensor information fusion technology in the field of mobile robot has become a hot spot in the field of Robotics problem. It combines the development of automatic control, artificial intelligence and decision making theory, and provides a kind of technical solution for the exploration of the unknown environment. Information fusion, also known as data fusion, began in 1970s. In 1983, the United States Defense Advanced Research Projects Agency launched the strategic computing program. The multi sensor information fusion is a major research project. In 1984, United States Department of defense set up a data fusion group, responsible for guidance, organization and coordination of the key technologies of the defense system research [1], which will be included in the defense of the 22 key technologies in In the military field, the application of information fusion, including the tactics and strategy of command, control, communication and intelligence has been realized in other fields. At present, the application of information fusion has been applied to other fields which refer to medical image processing and diagnosis system, biological information science, testing system and fault diagnosis system, weather forecast, modern manufacturing industry, intelligent transportation, intelligent robot and so on. Radio frequency identification, which refers to the RFID technology, is a typical radio frequency automatic identification technology that uses radio frequency signals to achieve recognition by a space called an alternating magnetic field or an electromagnetic field, in order to obtain the identification accuracy, high performance, large storage information, oil washing, etc., which is especially suitable for automatic recognition of the extreme work environment. The use of labels to replace the bar code and other products, can effectively complete the automatic management of storage, automatic collection, processing and information report for the realization of the goods information [2]. Figure 1 (a) - (b) shows the application of RFID technology in mobile robots. a. Mobile Robot Based on RFID Technology b. Robot Control And Positioning System Based On RFID Figure 1 Application of RFID Technology in Mobile Robot DOI /IJSSST.a.16.4B ISSN: x online, print

2 II. ROBOT MOTION CONTROL Robot motion control is accomplished by collecting all kinds of sensor information. Control of individual motor. The acquisition of various types of sensor information such as camera, ultrasonic sensor, electronic compass, encoding sensor and photosensitive sensor, etc.. This paper is based on the remote reading and penetration of non metallic materials and other characteristics to the robot's motion control. In this case, the data acquisition and processing of a sensor as a robot. Tags for the robot's external information, the robot through the reader to collect the information of the various tags, to get more accurate control. At present, most of the robot will move the motion control and sensing information acquisition as the two independent units to achieve, that is, a motor drive plate control motor, a sensor acquisition board to complete the signal acquisition. In order to meet the special requirements of the small mobile robot, the system is in the process of implementation, the sensor acquisition and robot motion control of the technology are synthesized in a fast control board, which can satisfy the general motion control and sensor information collection, can also be some simple behavior control such as steering, etc. on the block control board to complete, the flexibility of robot control is improved, and the occupancy of the controller is reduced, at the same time, it can make the data accumulation for the robot in the back, and save the resource of the operation [3]. III. BASED ON RFID THAT IS THE BASIC ALGORITHM A. Indoor Wireless Communication Mechanism Radio communication has many different mechanisms, which mainly depend on the radio transmitting power. For the RFID system in the interior, the main way of electric wave propagation is the visual line and non line of sight. The main contribution to the signal from the visual range, if the receiver is outside the range of sight, the transmitter transmits the signal energy to the receiver by means of transmission, reflection and refraction. And the important characteristic parameters, such as the received signal power and the radio propagation path loss, it is through the propagation model of wave reflection, wave diffraction and wave scattering propagation mechanism based on estimation [4]. Reflection: when the electromagnetic wave in the propagation process encountered in the size of the obstacle when the electromagnetic wave grow a lot of, reflection. When the wave is incident on the surface of the ground, buildings, and walls, reflection mode is the main mode of electric wave transmission. Reflection of an object's surface or floor, the reflected wave has great interference to the receiver. Diffraction: when the propagation path on the obstacle with sharp edges on the surface, happen diffraction phenomenon. At this time, occurrence of diffraction wave propagation will continue to encounter obstacles after wave propagation direction, equivalent to the wave propagation in the process of a "bending" phenomenon. In many cases, although there is no line of sight path, can also receive the signal, it is because of the existence of the diffraction propagation mode. In system working frequency band, mode of transmission diffraction also depends on the size of the obstacle, and the nature of waves, such as amplitude polarization method, etc.. Scatter: if there are many obstacles in the path of the radio propagation path, which are similar to the wave length or less than the wavelength of the wave. The scattering propagation mode is caused by the rough surface of the size of the obstacle. Scattering and diffraction obey the same physical theory, because of the energy emitted by the transmitter to many different directions. Diffraction is the three mechanism of the most complex, the most difficult to predict [5]. B. Channel Characteristics and Research Methods Indoor wireless environment is characterized by a small transmission power, more close to the coverage of the environment, the greater the change in the environment. For different buildings, indoor layout, material structure, building size and application types and other factors, which makes the communication environment has been a big difference. Even in different locations within the same building, the spread of the bad environment is not the same, and even a great difference. Therefore, the propagation of electromagnetic wave in the indoor environment is influenced by many factors, more complex than the spread in the outdoor environment [6]. The internal structure of the building will cause reflection, radio wave diffraction, refraction and scattering, this is also caused by the transmission of the signal through more than one way to reach the receiving end, the formation of multipath phenomenon. When the narrow band emission is caused by the fluctuation of the received signal envelope and phase of the received signal, the result is a series of delayed and attenuated pulse echo. Multipath fading greatly reduces the performance of indoor communication systems, however, the multipath is not controlled. In a clear understanding of the characteristics of the multipath medium, the transmitter and receiver can be designed to reduce the impact of multipath, therefore, the research on the characteristics of indoor wireless channel is the key to the successful design of communication system. The method of indoor wireless communication is directly dependent on the characteristics and the intrinsic properties of the communication system. In the study of indoor radio propagation, the following factors are considered: 1) The type of external wall, such as steel structure, glass, brick, concrete, a shield, room layout, decoration, etc.. 2) Because of the difference in the environment of the city and the suburbs; 3) Because of the difference between the structure and the inner city and the suburbs of the city; 4) The population density of the building has an important influence on the statistical properties of the received signal; 5) The transmission of the transmission, such as the transmitter and receiver at the same time in the indoor or outdoor; 6) The difference between indoor and outdoor transmitter height [7]. DOI /IJSSST.a.16.4B ISSN: x online, print

3 C. Commonly Used Positioning Algorithm In the radio frequency positioning system, in order to eliminate the error of different reader to locate the result, general use of relative positioning method. Select one of the reader as a service reader, then investigate the distance between the tags and the reader. Once the measurement values are obtained for the distance between the two reader, can be obtained from the distance between the two reader to locate the reader, a set of hyperbolic equations for the position of the position of the tag is obtained by the multiple reader to the position of the label. Let ( x, y ) be the estimated position of the label to be positioned, and the (X i, Y i ) is the known position of the first reader, depending on the geometry, the distance between the tag and the i reader is: 2 2 Ri (X i x) (Y i y) (1) The Ri 1 indicates that the distance between the reader and the reader is poor, and the distance between the reader and the reader is R i 1 R i R 1 (X i x) (Y i y) (X 1 x) (Y 1 y) (2) In order to solve the nonlinear equations, we can deal with it first. After proper mathematical transformation, it can be got: 2 Ri 1 2Ri 1R1 Ki 2Xi 1x 2Yi 1y K1 (3) Among them, Xi-1 Xi X1 (4) Yi-1 Yi Y1 (5) 2 2 Ki Xi Yi (6) The x, y and R 1 are considered as unknowns, and the upper type is a linear equation group [8]. D. Node Location Based on Node Location Based on the basic principle of RSSI positioning, the emission signal intensity of the transmitter is known, the receiving node calculates the transmission loss of the signal according to the intensity of the received signal, the transmission loss is converted to a distance by using theoretical or empirical models, the location of the nodes is calculated by using the algorithm. Radio propagation path loss has a great influence on the positioning accuracy of RSSI positioning algorithm. The common path loss model is free space propagation model, log distance path loss model, Hata model, log normal distribution model, and so on. For field application environment, the free space propagation model and the log normal distribution model are generally used to analyze and simulate. Free space radio propagation path loss model: Loss nlg( d0 ) 10nlg( f ) (7) Among them, d 0 is the distance from the source, the unit is km; f is the frequency, the unit is MHz; n is path attenuation factor [9]. IV. THE COMPOSITION AND BASIC WORKING PRINCIPLE OF RFID The RFID consists of three parts, including tags, readers, and computer communication networks, As shown in Figure 2. Figure 2 RFID System Diagram Tags: electronic tags are usually placed in the need to be reader can be connected to a computer communication identified, and the related information is stored. The reader network, so that data transmission and communication can read this information in a non-contact way. That is the functions are realized. Its work flow is as follows: after core part of RFID, the label is divided into two kinds of launching an antenna, the radio carrier signal is transmitted active tags and passive tags. Reader: the reader is a device to the reader; In the launch area, when an electronic tag is that uses radio frequency technology to read the relevant read, it can be activated, is a series of operations on this information about the item in a non - contact way. Readers signal, modulation, decoding, decryption, then the signal is can also write information to an electronic tag [10]. judged by its command request, password, permissions and Computer communication network: the function of the data so on; Then the antenna will be transmitted within the tag management and communication. A standard interface information code; The reader receives the signal of the tag DOI /IJSSST.a.16.4B ISSN: x online, print

4 and the signal is decoded by the receiver, send to computer; computer through judgment, the corresponding processing and control, the whole work is the process of each job through computer network communication connection, form an information platform, according to different requirements of the design of the corresponding software, so as to complete the corresponding function [11]. V. APPLICATION OF INFORMATION FUSION TECHNOLOGY IN MOBILE ROBOT LOCALIZATION SYSTEM In recent years, with the development of science and technology, computer technology, multi sensor information fusion technology has been developed rapidly. The meaning of information fusion can be summarized as a comprehensive treatment of multi-source or information from multiple sensors, to get more reliable and accurate conclusions. Multi sensor information fusion technology was proposed in 1970s, the technology is firstly applied in military field, which mainly includes the location, tracking and recognition of military target, specific applications include marine surveillance to ship, aircraft, submarines, land and sea and other platforms, for the detection, tracking and identification of targets and time, monitoring space can reach hundreds of miles of air, water, water, etc., air to air or ground to an air defense system, such as land, detection, tracking and identification of aircraft, surveillance space can reach hundreds of miles or miles range. At present, multi sensor information fusion technology is widely us ed in the field of civil [12-14]. Specific application areas are: image fusion field, industrial control field, intelligent robot, remote sensing, fault diagnosis and other fields. In recent years, both military and non military fields, fusion of multi sensor data has received wide attention. One reason is that, after data fusion technology, the relevant information and data from multiple sensors are integrated, compared with the use of isolated, single sensor, can get more accurate and detailed conclusions. Compared with the single sensor, multi sensor data fusion has several obvious advantages: First, multiple sensors are used simultaneously, comprehensive use of these observations by increasing the number of independent observations, can further improve the estimation of target velocity and position. If the data is integrated in one of the best ways, it will be able to receive the benefit of statistical significance; Second, the relative position or motion of a plurality of sensors can be utilized to improve the observation process. Third, using multi sensor can improve the observation ability, broaden the physical observation method can greatly improve the target observation ability. Application of multi sensor information fusion technology in mobile robot system, mainly will have redundancy fusion and synthesis to improve the accuracy and robustness of the whole positioning or navigation system with multiple sensor data [15]. Since 1980s, the technology has been used in mobile robot system. LASS Laboratory of France, using multi sensor information fusion technology to build a machine HILARE. The robot is composed of a vision sensor, a laser range finder and an ultrasonic sensor, information fusion algorithm using weighted average method. The environmental adaptability of the robot HILARE is greatly enhanced by the multi sensor information fusion technology, stable work in the environment. In 1997, Sojna robot successfully landed on mars, the robot uses dead reckoning method and extended Calman filter algorithm based on speed sensor, sensors such as sonar and mileage meter, achieved a good application of information fusion technology in mobile robot. In 1999, Sasiadek machine is used to fuse the information of INS/GPS with Calman filter technology. In 2001, the Robin robot using neural network fusion technology will laser sensor, tactile sensor and ultrasonic sensor information fusion. Asimo humanoid robot developed by Honda Corporation of Japan (ASIMO), is currently the most advanced humanoid walking robot. It contained a large number of sensors, including visual sensor, infrared sensor, CCD camera and ultrasonic sensor, data fusion of these sensors can be performed by human recognition, avoiding obstacles and other operations. The U.S. Defense Advanced Research Projects Agency (DARPA) proposed an autonomous ground vehicle (ALV) program, in 1984, ALV research started, to 1989 has achieved a lot of results, ALV using sonar, laser ranging sensor and camera sensor. The research progress of the application of in mobile robot is greatly promoted by the intervention of DARPA. Design and manufacture of autonomous land vehicles by Marietta Martin, in the application, Determine the height and tilt angle of the sonar using sonar, using a color camera to find the obstacles in the environment, the use of laser ranging sensor positioning. In summary, in the process of robot work, the use of multi sensor information fusion, to be far superior to the use of a single sensor, so, research on multi sensor information fusion technology has become one of the important topics in the field of robotics research. VI. ROBOT PARAMETER CALIBRATION SYSTEM DESIGN BASED ON RFID The development of parameter calibration system is in order to move the robot in front of the indoor stadium, get active RFID the environmental parameters and path loss index in the system parameters of the location, provide a convenient tool for the operator. The parameters calibration system flow is shown in Figure 3. Figure 3 Flow of Parameter Calibrated System DOI /IJSSST.a.16.4B ISSN: x online, print

5 VII. MOBILE ROBOT LOCALIZATION EXPERIMENT Aiming at the positioning method based on, in this section, the proposed method is verified by experiments. In a 5.8 meter long, 5.8 meter wide laboratory environment, based on the communication range of the nodes, after many repeated tests, the author concluded that only 4 beacon nodes are required to meet the requirements of indoor positioning. The coordinates of the 4 beacon nodes are, (0, 0), (0, 2), (2, 2), and (2, 0), the mobile robot is equipped with a node that is to move the node. The mobile node to the beacon node to initiate the handshake protocol, a star shaped network with mobile node as the center of the handshake is successful. Signal strength value (RSSI) for the mobile node to receive the beacon node, by converting it into distance, the coordinates of the mobile node can be estimated by the beacon node coordinates and the weight method. Star positioning network stability, the RSSI values of the beacon nodes received by the mobile node at different positions are recorded, in order to improve the calculation accuracy, the position of mobile robot is changed 5 times in this experiment, at each position, the RSSI value of each beacon node is read 10 times, mean value is used as the RSSI value of the conversion distance operation. Positioning experimental data as shown in Table 1. TABLE 1 EXPERIMENTAL DATA LIST Actual coordinates (unit: m) RSSI-1 RSSI-2 RSSI-3 RSSI-4 Calculated coordinates (unit: m) Mean square error (0.5,0.5) (0.6,0.57) (0.5,1) (0.76,0.95) (1,1) (0.80,0.97) (1,3) (1.22,2.76) (3,1.5) (3.17,1.21) (5,4.5) (5.31,4.20) Mean error(m) From the table of data can be seen, the node positioning accuracy is about 20cm, in comparison with the indoor environment of mobile robot positioning needs. Of course, the positioning accuracy is improved, we can improve the accuracy of positioning by reducing the interference factor, increasing the number of beacon nodes or changing the location of the beacon nodes. VIII. CONCLUSION This paper introduced the related theory of mobile robot localization technology, including RFID technology and information fusion technology, and based on this, the corresponding system model was established. The mobile robot has good automatic positioning performance by the experimental simulation based on RFID and information fusion technology. The technology has broad development space and application prospect. ACKNOWLEDGEMENTS This work is supported by 2015 Science and Technology Research Project of Chongqing Academic Affairs, Intelligent Electronic Guiding System Based on NFC and RFID Technologies. Project Number: KJ REFERENCES [1] Betke M, Gurvits L. Mobile robot localization using landmarks. Robotics and Automation, IEEE Transactions on, vol. 13, no. 2, pp , [2] Curtin J, Kauffman R J, Riggins F J. Making the MOST out of RFID technology: a research agenda for the study of the adoption, usage and impact of RFID. Information Technology and Management, vol. 8, no. 2, pp , [3] Dutta A, Lee H L, Whang S. RFID and operations management: technology, value, and incentives. Production and Operations Management, vol. 16, no. 5, pp ,2007. [4] Fierro R, Lewis F L. Control of a nonholonomic mobile robot using neural networks. Neural Networks, IEEE Transactions on, vol. 9, no. 4, pp , [5] Hossain M M, Prybutok V R. Consumer acceptance of RFID technology: An exploratory study. Engineering Management, IEEE Transactions on, vol. 55, no.2, pp , [6] Jones P, Clarke-Hill C, Hillier D, et al. The benefits, challenges and impacts of radio frequency identification technology (RFID) for retailers in the UK. Marketing Intelligence & Planning, vol. 23, no. 4, pp , [7] Kosaka A, Kak A C. Fast vision-guided mobile robot navigation using model-based reasoning and prediction of uncertainties. CVGIP: Image understanding, vol. 56, no. 3, pp , [8] Leonard J J, Durrant-Whyte H F. Mobile robot localization by tracking geometric beacons. Robotics and Automation, IEEE Transactions on, vol. 7, no. 3, pp ,1991. [9] Nath B, Reynolds F, Want R. RFID technology and applications. IEEE Pervasive Computing, vol. 5, no. 1, pp.22-24,2006. [10] Ngai E W T, Cheng T C E, Au S. Mobile commerce integrated with RFID technology in a container depot. Decision Support Systems, vol. 43, no. 1, pp.62-76,2007. [11] Opasjumruskit K, Thanthipwan T, Sathusen O. Self-powered wireless temperature sensors exploit RFID technology. Pervasive computing, IEEE, vol. 5, no. 1, pp.54-61, [12] Se S, Lowe D, Little J. Mobile robot localization and mapping with uncertainty using scale-invariant visual landmarks. The international Journal of robotics Research, vol. 21, no. 8, pp , [13] Ustundag A, Tanyas M. The impacts of radio frequency identification (RFID) technology on supply chain costs. Transportation Research Part E: Logistics and Transportation Review, vol. 45, no. 1, pp.29-38, [14] Wamba S F, Lefebvre L A, Bendavid Y, et al. Exploring the impact of RFID technology and the EPC network on mobile B2B ecommerce: A case study in the retail industry. International Journal of Production Economics, vol. 112, no. 2, pp ,2008. [15] Want R. An introduction to RFID technology. IEEE Pervasive Computing, vol. 5, no. 1, pp.25-33,2006. DOI /IJSSST.a.16.4B ISSN: x online, print