资源描述
外文文献翻译在导盲应用中的超声波动态距离修正系统 Robert X.Gao,Member,IEEE and Chuan Li Institute for micromanufacturing Louisiana Tech University P.O.Box 10348 Ruston,LA 71272 USA摘要：一个基于以微处理器为基础的超声波距离修正系统已被设计出来，它的两个管脚之间用类似PVC管来封装。此系统，作为盲人导盲应用，它管脚里面包含四个超声波传感器做为一个整体，外加一个微处理器从而构成了这个系统。它可以实时提供使用者前进的道路上最大大约5米的距离内所遇到的障碍物的距离和高度信息。在实验室进行各种移动模式的测试中，在5米的范围内最大的相关测量误差低于14%。当测量范围为3米时，这个误差还会降低到5%。把此系统整体小型化后，可以做为盲人的日常导盲的应用。关键词：盲人导盲，导盲棒，传感器集成，设备小型化1 绪论过去的十年表明，医学工程已经取得了巨大的进步。医疗设备的商业化有着宽广的前景，它能为残疾人充分的提供一个更为舒适的生活环境。然后，科技的进步，似乎还没对电子导盲（ETAs）领域产生太多的影响。虽然一个ETAs产品可以造就一个行业，但是，盲人群体可以接受的费用是低的令人吃惊。即使是现在，大部分的盲人还是在选用传统的拐杖来作为自己首选的出行辅助工具，它可以提供给使用者一个半径约为1米的弧形的延伸空间来感觉是否有障碍物。但是，使用拐杖来作为盲人的辅助工具，它有有不可克服的局限性，它不能为使用者提供腰至头部的保护。因此，使用拐杖作为盲人的出行辅助工具，它潜在的危险非常的高。近年来，微电子电路的高速发展，这为拐杖植入嵌入式微传感器提供了机会，它可以在传统的外观和功能上，再增加额外的功能，使基于传统的拐杖上，功能变的更为完善，它可以为使用者提供更为宽广的障碍物提示报警功能。在这篇论文中，将呈现出上述盲人导盲棒的设计和测试。实验室的测试实验包含有不同的典型的障碍物和导盲棒不同的移动方式测试，用于提高传感器整体化导盲棒的适用性，以求能为盲人的日常出行提供方便。盲人行走中所产生的不足之处仍然很多。目前并没广泛接受ETA可能是因为没有在心理上接受TEA设计者组织和制造商，缺乏吸引潜在用户，或者是因为目前发展的现有技术还不能实现一个复杂并且有成本效益，可依赖的人类环境改造学产品。2 原理障碍物的测量是基于超声波的回声测量。如下图1是障碍物与四个（Tr1Tr4）一体化超声波变换器的的方位模型。传感器Trl，工作频率40KHz，是配置一个发射器和一个接收器，Trl仅作为一个接收器。障碍物与手杖持有人的水平距离（D），以及手杖的垂直高度（H）可以由下面式子得到：信号传输路径L1和L2都取决于测量周期的开始到接收到Tr1和Tr2的回声的传播时间，参数1-4通过变换器设置决定。手杖与地之间的角度和手杖顶端到地面的距离L4，考虑由于手杖在实际中运动情况造成的测量误差，由辅助传感器Tr3和Tr4决定。图1 障碍物的识别3 试验为了证明传感器集成技术，实验室用三个典型的对象来引证：1）直径为20cm的无形花盆，2）截面为10mm×10mm杆长为1.7m的木杆和3）一个40毫米的钢球。这些对象被挂在一个离地面高度约1.7米，到手杖的距离为5米内。两个超声波测距模块用作Tr1和Tr2。对于Tr3和Tr4，两个模块在马萨产品中得到应用。这些与1.5长PVC管和5厘米的二步格成为一体。数据的采集和处理协调由摩托罗拉的微处理器M68HC711E9控制（图2）。高度和距离信息是通过从5米每远渐渐移动手杖到挂着的对象获得的，每25厘米采集一次数据。为了模拟手杖的实际使用，手杖作为接近的对象，它经历了三种类型的议案：手杖末端放到地面时，离末端15厘米左右的电弧，离末端5厘米的助听器，两个成为一个组合。所获得的数据再与用工业尺进行比较。如图3所示，高度测量的最大的相对误差小于14%；对于远距离的测量，测量误差在12%以内。测量精度随着用户接近障碍而迅速提高：在3米以内的距离，误差迅速下降到小于5%，这意味着一个更好的保护。图2 系统构造图3远距测量和高度测量的误差4 结论手杖用户面前，超声波传感器阵列在局部移动的路径障碍里是有效的。测量精度的提高可以通过系统的修改和应用误差补偿技术。今后的工作包括硬件和软件的优化，包括其他对象的领域测试和使用微制造技术的最小化系统。A DYNAMIC ULTRASONIC RANGING SYSTEMAS A MOBILITY AID FOR THE BLINDRobert X. Gao, Member, IEEE and Chum LiInstitute for MicromanufacturingLouisiana Tech UniversityP.O. Box 10348Ruston, LA 71272 USAAbstract - A microprocessor-based ultrasonic ranging system, configured around a PVC tube resembling the form of a long cane, has been designed. The system, intended to serve as a mobility aid for the blind, consists of four ultrasonic sensors integrated into the cane, and a microcontroller.It is able to provide real-time information on the distance and height of over-hanging obstacles within 5 meters along the travel path ahead of the user. Lab experiments under various cane motion patterns have shown that at 5 meters, the maximum relative error of the measurement is below 14%. This drops to less than 5%, when the measurement range is within 3 meters. Uponminiaturization, the system can be integrated into a conventional long cane for practical daily use by the blind.(Keywords: mobility aid for the blind, long cane, sensorintegration, device miniaturization)I. INTRODUCTIONThe past decades have witnessed tremendous technological progress in biomedical engineering. A wide range of rehabilitation equipments have been commercialized, contributing Substantially to creating a more comfortable living environment for the disabled. Nevertheless, in the field of electronic travel aids (ETAs) for the blind, the research efforts seem to have made less impact. Although a variety of ETAs have been made commercially available l-31, their acceptance rate by the blind community remains surprisingly low 4. Up until now, the majority of blind are still using the traditional long cane as their primary travel aid, which provides an extended spatial sensing within an arc of the shoulders width about 1 m ahead of the user. However, due to its inherent limitation, the use of a long cane does not provide protection for the user between hisher waist and head. Thus, the potential for injuries arising from blind walking remains very high. The lack of a widely acceptable ETA so far may either be due to the lack of understanding of the psychological aspects of the blind community from the ETA designers and manufacturers and subsequently, the lack of appeal of the products to the potential users, or its because the technology available at the time of the development was not suflicient to enable a sophisticated yet cost-effective, reliabde and ergonomic product.The rapid development of the emerging microelectromechanical systems (MEMS) technologies in recent years has opened up opportunities for introducing a new type of microsensor-embedded long cane which, while maintaining the same appearance and function %1s its Wtional counterpart, is capable of perceiving overhangiig obstacles outside the detectable range ofthe conventional canes, thus providing a substantiallybetter protection of its users. In this paper, the design and performance tests of aprototype of such a system is presented. Laboratory experiments; involving different representative obstacles and various cane motion patterns have verified the applicability of such a sensor-integrated long cane for thepractical daily use by the blind people.11. PRINCIPLEThe obstacle detection is based on the echo-time measurement. Schematically shown in Fig. l is an instantaneous position of the prototype cane with four integrated ultrasonic transducers T, - T,. Transducer Trl, working at 40 1<Hz, is configured both as a transmitter and a receiver whereas Tfl works only as a receiver. The horizontal distance (D) of an obstacle (OBJ) to the cane holder and its vertical height 0 can be given as:The signal trave1,pziths L, and L, are determined by the time elapsed between the start of the measurement cycle and the receipt of the echos from T, and Ta. Parameters 6, - S, are given by the transducer setup. The angle ct between the cane and the ground and the distance L, from the cane tip to the ground take into account the measurement errors caused by the cane motion in a realistic situation, and are determined by the auxiliary transducers Tr3 and T,.III. EXPERIMENTSTo veri the sensor-integration technique, lab experiments were conducted using three representative objects: 1) aplastic flowerpot about 20 cm in diameter, 2) a 1.7 m long wood rod with a cross section of 10 mm x 10 mm, and 3) a 40 mm steel ball. These objects were hung at a height of about 1.7 meters above the ground, within a distance of about 5 m from the cane. Two Polaroid ultrasonicrsfnging modules were used for T, and Ta. For TI:,a nd TF(t,w o modules from the Massa Products were applied. These were integrated into a PVC tube about 1.5 m long and 5 cm in dimeter. The coordination of the data acquisition and processing were controlled by a Motorolamicroprocessor M68HC711E9 (Fig. 2). The height and distance information were obtained by moving the cane from a 5 m distance gradually to the hanging object, with the data taken at an interval of every 25 cm. To simulate the realistic cane use, as the cane approached to the object, it underwent three types of motions: arcing of the cane tip of about 15 cm to the left and right, tapping of the cane tip on the ground, with the tip lift being about 5 cm, and a combination of the two. The data obtained were then compared with the measurements using an industrial ruler. As shown in Fig. 3, the maximum relative error of the height measurement is less than 14%; for the distance measurement, it's within 12%.The measurement accuracy increases rapidly as the user moves closer to the obstacle: within a range of 3 meters, the error drops quickly to less than 5%, meaning a better protection.IV. CONCLUSIONSThe ultrasonic sensor array is effective in localizing obstacles within the travel path in front ofthe cane user. The measurement accuracy can be increased through system modification and applying error compensation techniques. Future work include hardware and software optimization, field tests involving othere objects, and system miniaturiration using various micromanufacturing techniques.ACKNOWLEDGEMENTThis research was supported by the National Science Foundation under grant BES-9402818. The authors thank M. Rasbury and J. Cook for fabricating the test devices.REFERENCES1 L. Kay, "Electronic Aids for Blind Persons: An InterdisciplinarySubject", IEE Proceedings, Vol. 131, No. 7, pp. 559-576, 1984.2 A. Dodds, J. Armstrong, and C. Shingledecker, "The Nottingham Obstacle Detector: Development and Eevaluation", J. of Visual and Blindness, Vol. 75, pp.203-206, May, 1981.3 A. Heyes, "The Sonic Pathfinder A New Electronic Travel Aid", J; of Visual Impairment and Blindness, Vol. 78,No. 5, pp. 200-202, May, 1984.4 M. Sanders and E. McComick, "Human Factors in Engineering and Design", 7th edition, McGraw-Hill, Inc.,1993.9
展开阅读全文