Affiliation:
1. AT&T Bell Laboratories Murray Hill, New Jersey 07974
Abstract
A generic range sensor is a member of a collection of sensors that provide surface location information. This collection includes ultrasonic, laser, structured light, and tactile sensors. The tactile sensor can be viewed as a degenerate range sensor when in contact with the surface. This paper will deal pri marily with the ultrasonic range sensor, which is probably the most difficult to handle, but extension of the techniques to other sensors is apparent. The use of ultrasonic range finding to gather data about object surfaces in an air transmission medium poses several problems. First, there is very little information returned in a single range measurement. One range measurement tells us that some surface exists, approximately orthogonal to the transmission axis of the transducer, at the measured range. However, the surface can lie anywhere within the sensitivity pattern of the transducer, and it can be oriented at any angle that reflects energy back to the transducer. Moving the trans ducer adaptively allows us to obtain much more information about the surface including determination of the actual path of the signal echo point on the surface of the object. The topic of ultrasonic echo trajectories introduced in an earlier paper will be reviewed. A property of ultrasonic echo trajectories on object surfaces is exploited. These trajectories are very nearly geodesic for the ultrasonic transducers being used when they are moved on a straight path parallel to a smooth regular surface. This property has several useful consequences. The surface path of the echo trajectory can be approximately determined. Fur thermore, surface normals can be obtained everywhere along this path. Thus, fundamental properties of the surface can be determined.
Subject
Applied Mathematics,Artificial Intelligence,Electrical and Electronic Engineering,Mechanical Engineering,Modelling and Simulation,Software
Reference30 articles.
1. Agin, G.J. and Binford, T.O. 1973, Stanford, CA. Computer description of curved objects. 3d Int. Joint Conf. Artificial Intelligence 629-640.
2. Brown, M.K. 1984, Atlanta, GA. Computer simulation of controlled impedance gripper. IEEE Int. Conf. Robotics. 442-450.
Cited by
18 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献