【導讀】本設(shè)計實現(xiàn)智能小車在避開障礙物的前提下對黑色軌跡循跡。采用紅外光電管模塊。芯片控制小車電機的速度及轉(zhuǎn)向，從而實現(xiàn)自動循跡避障的功能。L298N驅(qū)動電路完成，速度由單片機輸出的PWM波控制。一次，比較小車左右兩側(cè)離障礙物的距離，選擇較開闊的方向行走，從而避開障礙物。宇航、國防等領(lǐng)域。近年來機器人的智能水平不斷提高，并且迅速地改變著人們的生活。人們在不斷探討、改造、認識自然的過程中，制造能替代人勞動的機器一直是人。動行走和駕駛的重要部件。視覺的典型應(yīng)用領(lǐng)域為自主式智能導航系統(tǒng)，對于視覺的各。通過大量的運算也只能識別一些結(jié)構(gòu)化環(huán)境簡單的目標。視覺傳感器的核心器件是攝像。管或CCD，目前的CCD已能做到自動聚焦。但CCD傳感器的價格、體積和使用方式上并

　　

【正文】 easured by sonar i (using its own time frame). Let DTi,j be the delay in transmitting the chirps between sonar i and j. Accordingly, DTi,i = 0, and Ti,j = DTj,i. Finally, let the total time of flight of the signal be denoted by TOFi,j = Ti,j + DTi,j. For example, the total time of flight of signal 1 (transmitted by sonar 1) while trave ling from sonar 1 to the reflector and back to sonar 1 is TOF1,1 = T1,1. The time of flight of signal 1 from sonar 1 to the reflector and back to sonar 2 is given by TOF2,1 = T2,1 + DT2,1. 28 Figure 4: A typical signal sequence of the system Once the TOF of signals from all sonars to all other sonars (in the case of 2 sonars there are 4 measurements 2 per signal), the position of reflector O can be calculated according to geometric interpolation between sonar 1 and sonar 2. Many researchers have investigated the use of geometric triangulation in multisonar systems [LeMay and Lamancusa 1992。 Kleeman and Kuc, 1995]. Typically these systems consist of an array of several sonars, but only one sonar transmits a signal at a time while all other sonars receive the echo. In addition to the range measurement, the geometric interpolation procedure provides an estimate of the bearing of the reflector [Nagashima and Yuta 1992], differentiating edges, corners and walls [Barshan and Kuc 1990。 Sabatini 1992], and radius of curvature [Peremans et al., 1993]. However, due to their limited transmission rate, these systems are unsuitable for fast moving vehicles. Figure 5 illustrates the geometric triangulation procedure. A coordinate system C1 is 29 attached to sonar 1, which is located at the origin of this system and transmits its coded chirps in the V1 direction (along the Yaxis). Sonar 2 is located at position P2 and transmits its coded chirps in the direction V2 creating angle q with the Xaxis. Reflector O is positioned at P3 (to be determined) within the lobe of the chirp transmitted by sonar 2. Assuming the echo transmitted by sonar 2 is received by both sonars, the location of reflector O is calculated by both sonars: The calculation for Sonar 2 is based on TOF2,2 and that for sonar 2 is based on TOF1,2. The uncertainty of the reflector position as measured by sonar 2 is within the circular arc (177。15o for conve ntional sonar) as shown in Fig. 5. The uncertainty of the reflector position as measured by sonar 1 is within an elliptic arc defined by the geometric relation between the two. The actual position of the object is accurately determined based on the intersection between the circular arc and the elliptical arc, as shown in Fig. 5. Figure 5: Positioning of the reflector based on measurements of sonar 1 and sonar 2. In addition to the accurate location of the reflector, the intersection between the two arcs provides information required for sorting the echoes detected by the sonars. When several echoes are perceived by a particular sonar, it is essential to relate each echo to echoes perceived by other sonars. However, some of the measurements can be erroneous due to reflection from several reflecting surfaces, sensor inaccuracies, or external noise. The 30 intersection method rejects this erroneous data, as only arcs that intersect within the lobe of the transmitted chirp represent acceptable echoes. Arcs that either do not intersect at all, or intersect outside the lobe pattern of the transmitted signal represent erroneous echoes. This solution provides a unique method for localizing reflectors ensures the rejection of most erroneous data. This phenomenon is demonstrated in experiment 2 of the next section. 3. Experimental Results This section provides results of basic experiments designed to test the feasibility of the proposed technique. First, consider the system shown in Fig. 5, in which sonar 1 is positioned at (0, 0) and sonar 2 at (50, 5), with q = 85o. The reflector is at (50, 100) close to the left edge of the lobe. In this experiment each sonar transmits one coded chirp. The measurements obtained by the two sonars are TOF1,2 = cm and TOF2,2 = cm (no echoes are perceived from the chirp transmitted by sonar 1). Assuming a conical propagation profile for the emitted chirp, the measurement of sonar 1 constructs an elliptical arc and sonar 2 a circular arc. The intersection of the two arcs establishes the position of the object to be (, ), an offset of cm from the actual position, as shown in Fig. 6. Figure 6: Experimental results for the configuration shown in Fig. 5. Next, the more plex configuration shown in Fig. 7a is tested. The sonars are positioned at the same locations as in the previous experiment, with three vertical cylindrical reflectors (5 cm diameter), positioned at (14, 100), (40, 300), and (95, 200). The sonars are 31 firing one coded chirp each, 5 ms apart (sonar 2 after sonar 1). The timing diagram in Fig. 7b shows when each echo was received by each sonar. As shown, sonar 1 receives its own echoes reflected from reflectors 1 and 2, as well as the echoes transmitted by sonar 2 and reflected from reflector 2. Sonar 2 receives its own echoes reflected from reflectors 2 and 3 and the echo transmitted by sonar 1 and reflected by reflector 2. Figure 7a: Set up for experiment 2. Figure 7b: Signals transmit/receive schedule for the set up of experiment 2. Table I summarizes the results of this experiment. The first column provides a serial number to each echo and the second column (R, T) indicates the sonar that transmitted the chirp (R) and the sonar that received that echo (T). The other columns provide time data on 32 receipt of the echo by each sonar and the total time of flight from transmission, through reflection and perception (TOFi,j). Finally, the translation of TOF to distance is shown, based on Vair, the speed of sound in air at