Frank Pipitone

SUMO: spacecraft for the universal modification of orbits

SUMO, or Spacecraft for the Universal Modification of Orbits, is a risk reduction program for an advanced servicing spacecraft sponsored by the Defense Advanced Research Projects Agency and executed by the Naval Center for Space Technology at the Naval Research Laboratory in Washington, DC. The purpose of the program is to demonstrate the integration of machine vision, robotics, mechanisms, and autonomous control algorithms to accomplish autonomous rendezvous and grapple of a variety of interfaces traceable to future spacecraft servicing operations. The laboratory demonstration is being implemented in NRL’s Proximity Operations Test Facility, which provides precise six degree of freedom motion control for both the servicer and customer spacecraft platforms. This paper will describe the conceptual design of the SUMO advanced servicing spacecraft, a concept for a near term low-cost flight demonstration, as well as plans and status for the laboratory demonstration. In addition, component requirements for the various spacecraft subsystems will be discussed.

Tripod operators for recognizing objects in range images: rapid rejection of library objects

The tripod operator, a class of feature extraction operators for range images which facilitate the recognition and localization of objects is described. It consists of three points in 3-space fixed at the vertices of an equilateral triangle and a procedure for making several scalar measurements in the coordinate frame of the triangle. The triangle is moved as a rigid body until the three vertices lie on the surface of some range image or modeled object. The resulting measurements are local shape features which are invariant under rigid motions. These features contain all the shape information in the surface points involved, and no other information. Tripod operators are applicable to all 3-D shapes. A cloud of points in feature space was generated for each object by random placement of the operator. Then new feature measurements were made by operator placements in a range image containing one of those objects. Using a simple nearest-neighbor approach, the authors determined which objects were rejected and which remained as recognition candidates. Experiments showed that tripod operators had excellent discriminating power.<<ETX>>

Three-dimensional computer vision for micro air vehicles

Results are described of an ongoing project whose goal is to provide advanced Computer Vision for small low flying autonomous aircraft. The work consists of two parts; range-based vision for object recognition and pose estimation, and monocular vision for navigation and collision avoidance. A wide variety of range imaging methods were considered for the former, and it was found that a promising approach is multi-ocular stereo with a pseudo-random texture projected with a xenon flash. This provides high range resolution despite motion, and can be small and light. The resulting range images, taken at a few meters range, would support the use of Tripod Operators, an efficient and general method for recognizing and localizing surface shapes in 6 DOF. This would provide the ability to recognize immediately upon encounter many kinds of targets. The monocular navigation system is based on finding corresponding features in successive images, and deducing from these the relative pose of the aircraft. Two methods are under development, based on horizon registration and point correspondences, respectively. The first can serve as a preprocessor for the second. This approach aims to continuously and accurately estimate the net motion of the vehicle.

An Artificial Intelligence Approach to Analog Systems Diagnosis

This paper describes some general diagnostic reasoning techniques which exploit recent advances in the field of artificial intelligence. They are applicable to a variety of human-engineered systems, including hydraulic, mechanical, and optical ones, but the primary focus has been on electronic systems. These techniques were developed over a period of several years of research in this area at the Naval Research Laboratory. One of the products of this research is a fully implemented diagnostic reasoning system called FIS which embodies these techniques and is in use at a variety of government and industrial laboratories.

Rapid recognition of freeform objects in noisy range images using tripod operators

The tripod operator is a class of feature extraction operators for range images. It facilitates the recognition and/or localization of objects by extracting a few sparse point samples in a regimented way; n sampled surface points yield only n-3 scalar features containing all the surface shape information in these points and no other information. They have complete six degrees-of-freedom (DOF) isometry invariance. Isolated objects are recognized in noisy range images using a statistical treatment of a few isolated tripod operator placements. Experiments show fast, reliable recognition.<<ETX>>

Efficient RPG detection in noisy 3D image data

We address the automatic detection of Ambush weapons such as rocket propelled grenades (RPGs) from range data which might be derived from multiple camera stereo with textured illumination or by other means. We describe our initial work in a new project involving the efficient acquisition of 3D scene data as well as discrete point invariant techniques to perform real time search for threats to a convoy. The shapes of the jump boundaries in the scene are exploited in this paper, rather than on-surface points, due to the large error typical of depth measurement at long range and the relatively high resolution obtainable in the transverse direction. We describe examples of the generation of a novel range-scaled chain code for detecting and matching jump boundaries.

Moving-correlation-code triangulation range imaging

A new method is described for obtaining accurate range images at high sped in a low-cost instrument. A prototype has been built and tested, and a patent application submitted. The method resembles grid-coding in that a camera and a stripe projector are directed at a scene, but the projector is different. It consists of a thin light source on the axis of a turntable, and a binary mask conforming to a cylinder coaxial with this. The mask has alternate black and clear stripes parallel to the axis. It forms a DeBruijn sequence, i.e., a sequence in which all possible sub- sequences of given length occur. No lens is used, deliberately smoothing the resulting illumination. In operation, the turntable rotates, and six consecutive images are taken at uniform intervals. A given pixel records six consecutive samples of a scene point. This six-vector, when normalized to unity to accommodate reflectance variations, is unique to the place in the sequence form which it came. Thus we can compute the position in 3-space of the surface point at which the pixel is looking. Observed accuracy is .1 millimeter at 30 centimeters range.

Tripod Operators for Efficient Search of Point Cloud Data for Known Surface Shapes

We address the problem of searching large amounts of 3D point set data for specific objects of interest, as characterized by their surface shape. Motivating applications include the detection of ambush weapons from a convoy and the search for objects of interest on the ground from an aircraft. Such data can occur in the form of relatively unstructured point sets or range images, and can be derived from a variety of sensors. We study here the performance of Tripod Operators (TOs) on synthetic range image data containing the shape of an oil drum; a cylinder with planar top. Tripod Operators are an efficient method of extracting coordinate invariant shape information from surface shape representations using discrete samples extracted in a specially constrained manner. They can be used in a variety of ways as components of a system which performs detection, recognition and localization of objects based on their surface shape. We present experimental results which characterize the approximate accuracy of detection of the test shape as a function of the accuracy of the surface shape data. This is motivated by the need for an estimate of the required accuracy of 3D surveillance data to enable detection of specific shapes.

Manifold Learning for Compression and Generalization of Euclidean Invariant Signatures of Surface Shapes

We introduce an approach to the efficient recognition of families of surface shapes in range images. This builds upon earlier work on Tripod Operators (TOs), a method for extracting small sets of N points from 3D surface data in a canonical way such that coordinate independent shape descriptions can be efficiently generated and compared. Using TOs, a specific surface shape generates a signature which is a manifold of dimension ≤ 3 in a feature space of dimension d = N - 3. A runtime application of a TO on surface data generates a d-vector whose distance from the signature manifold is closely related to the likelihood of a match. Ordnance identification is a motivating application. In order to use TOs for recognizing objects from large sets of known shapes, and families of shapes, we introduce the use of manifold learning to represent the signature manifolds with piecewise analytic descriptions instead of discrete point sets. We consider the example of generalizing the signatures of several artillery shells which are qualitatively the same in shape, but metrically different. This can yield a signature that is only slightly more complex than the originals, but enables efficient recognition of a continuous family of shapes.

A Structured Light Range Imaging System Using a Moving Correlation Code

We describe a system which produces a dense accurate range image using 8 consecutive frames of camera data in conjunction with a special projector. A prototype has been built and tested, and yields a typical range error of about 0.2 mm at 2 meters range with a baseline just over one meter. The camera is directed at a scene, along with a stripe projector consisting of a thin light source (xenon tube and slit) on the axis of a turntable, and a binary mask conforming to a cylinder coaxial with this. The mask has alternate opaque and transparent stripes parallel to the axis. It forms a sequence in which each subsequence of given length n (8 here) is different. No lens is used in the projector, deliberately smoothing the resulting illumination in a shadowing process. In operation, the turntable rotates, and images are taken at uniform angular intervals for several consecutive frames. In the consecutive frames, a given pixel records samples of the brightness of a fixed surface point. The vector consisting of those samples, when normalized, is unique to the place in the sequence from which it came, thus enabling the computation, via a fast indexing process, of the 3D position of the surface point. The code is similar to a DeBruijn sequence, but modified to reduce range error. Several types of calibration and error compensation were used to produce an accurate range image.