Abstract—One of the methods for detecting the target position error in the laser tracking systems is using Four Quadrant (4Q) detectors. If the coordinates of the target center is yielded through the usual relations of the detector outputs, the results will be nonlinear, dependent on the shape, target size and its position on the detector screen. In this paper we have designed an algorithm with using neural network that coordinates of the target center in laser tracking systems is calculated by using detector outputs obtained from visual modeling. With this method, the results except from the part related to the detector intrinsic limitation, are linear and dependent from the shape and target size. Keywords—four quadrant detector, laser tracking system, rangefinder, tracking sensor I. INTRODUCTION HE laser tracking is a very accurate, noncontact system with the dynamic measuring ability that the broad performance confines, high sampling rate and automatic tracking are of its features [1]-[3]. With the development and improvement of the performance quality of the laser tracking systems, it is possible to use them in robots calibration, 3D measurements and controlling applications of automatic machines [1]-[4]. The method presented in this paper is one of the new methods used in tracking systems for movement and changing the direction to the desired position and recording it [2], [4]. This system benefits from a phase rangefinder, 2-axes tracking mechanism, 4Q detector, gimbal and sheet reflector [4]-[7]. In order to determine the target position, the laser tracking system uses distance information and the angel yielded in order from the rangefinder and the tracking mirror parts [4], [5]. The tracking is completed with measuring the amount of tracking error signal and alignment of the tracking mirror angels on the minimum of these errors [4]. But if the spot light reflected on the detector screen does not hit each of the four quadrants or if it is situated on the gap band, the detector does not give the desired results [8]-[11]. If the spot light reaches A. Salmanpour was with Electrical Engineering Department of AZAD University-Lahijan Branch. He is now with Guilan Power Distribution Company, Rasht, Iran (phone: +98-131-6665001; fax: +98-131-6664499; e-mail: ar.salmanpour@yahoo.com). Sh. Mohammad Nejad is with Iran University of Science and Technology, Narmak, Tehran, Iran. (phone: +98-21-77240540; fax: +98-21-77240540; e-mail: shahramm@iust.ac.ir).He is a Member of IEE, and a Certificated Chartered Engineer (CENG) of IET. the edge of the detector, some errors occur in laser tracking system while determining the target position [8]-[11]. In this paper, through using artificial neural network, an algorithm is presented that calculates the target position without being dependent on these limitations. II. LASER TRACKING SYSTEMS A. Performance Analysis In this system, there are tracking sensors and rangefinders separated from each other. Fig.1 shows the tracking system configuration. At first, the sent signal from the laser through the tracking mirrors is directed to the target on which sheet reflector is installed. If the radiation does not hit the center of the reflector, there will be offset that indicates tracking error (the lateral displacement) [4]. Fig. 1 The tracking system configuration The amount of tracking error and distance is determined by the returned radiation from the target in 4Q detector and rangefinder parts after passing through the tracking system. The target tracking is done by using the error tracking signals x Δ and y Δ extracted from the 4Q detector. These signals are used for calculating the correct angels of the tracking axes in order that the sent radiation hits the target center. Then the signals are sent to the motor drivers in order to align the tracking mirrors angels on this value. The signals needed for tracking system correction that should be sent to the motor drivers are yielded though these relations [4]: β α 2 cos . 2 l y s Δ ≅ Δ (1) A. Salmanpour, Sh. Mohammad Nejad The Performance Improvement of the Target Position Determining System in Laser Tracking Based on 4Q Detector using Neural Network T World Academy of Science, Engineering and Technology International Journal of Electronics and Communication Engineering Vol:5, No:8, 2011 966 International Scholarly and Scientific Research & Innovation 5(8) 2011 scholar.waset.org/1307-6892/10347 International Science Index, Electronics and Communication Engineering Vol:5, No:8, 2011 waset.org/Publication/10347