Three studies that broadly relate to the issue of sensing for intelligent robotic are reviewed. The first involves the data-driven construction of a dynamic world model for use in mobile robot navigation. The second d...
详细信息
Three studies that broadly relate to the issue of sensing for intelligent robotic are reviewed. The first involves the data-driven construction of a dynamic world model for use in mobile robot navigation. The second demonstrates a machine learned, rule-based controller for a pole and cart problem. The third, for which results are provided, is a sensory integration exercise involving a vision and a taction system embedded in a multirobot work-cell.< >
Developed herein is a formal theory for stereo vision which unifies existing stereo methods and predicts a large variety of stereo methods not yet explored. The notion of "stereo" is redefined using terms wh...
详细信息
Developed herein is a formal theory for stereo vision which unifies existing stereo methods and predicts a large variety of stereo methods not yet explored. The notion of "stereo" is redefined using terms which are both general and precise giving stereo vision a broader and more rigorous foundation. The variations in imaging geometry between successive images used in parallax stereo and conventional photometric stereo techniques are extended to stereo techniques which involve variations of arbitrary sets of physical imaging parameters. Physical measurement of visual object features is defined in terms of solution loci in feature space arising from constraint equations that model the physical laws that relate the object feature to specific image features. Ambiguity in physical measurement results from a solution locus which is a subset of feature space larger than a single measurement point. Stereo methods attempt to optimally reduce ambiguity of physical measurement by intersecting solution loci obtained from successive images. A number of examples of generalized stereo techniques are presented. This new conception of stereo vision offers a new perspective on many areas of computervision including areas that have not been previously associated with stereo vision (e.g. color imagery). As the central focus of generalized stereo vision methods is on measurement ambiguity mathematical developments are presented that characterize the "size" of measurement ambiguity as well as the conditions under which disambiguation of a solution locus takes place. The dimension of measurement ambiguity at a solution point is defined using the structure of a differentiable manifold and an upper bound is established using the Implicit Function theorem. Inspired by the Erlanger program of F. Klein generalized stereo methods are equivalently described by the algebraic interaction of the symmetry group of automorphisms (i.e. bijections) of feature space into itself leaving a measurement so
A four-term project carried out with students of computer science at the Univ. of Bremen, West Germany, as part of the degree requirements is described. The project is called ROMEO, which is a German acronym for '...
详细信息
A four-term project carried out with students of computer science at the Univ. of Bremen, West Germany, as part of the degree requirements is described. The project is called ROMEO, which is a German acronym for 'robots and people: an occlusion'. The subject selected for the project is a version of the classic hand-eye problem of robotics: given a disassembled 3-D puzzle under the observation area of a small TV camera, the system to be implemented must identify the parts and determine their reference coordinates, as well as the degree of 2-D or 3-D rotations, and control a robot arm to assemble the puzzle. A term-by-term outline of ROMEO is given including both the material presented to the students, their organization into working groups, and results.< >
A heuristic approach for the normalization of the unevenness of digital arcs is proposed. This procedure is based on the development of a class of progressive windows that attempt to cover the existing unevenness. An ...
详细信息
A heuristic approach for the normalization of the unevenness of digital arcs is proposed. This procedure is based on the development of a class of progressive windows that attempt to cover the existing unevenness. An algorithm, HRA, and a number of illustrative examples are also provided. This work is a part of the project KALYPSO (a hierarchical picture recognition language) used on the HERMES vision machine. An additional application of the HRA algorithmic scheme could be the recognition and normalization of integrated circuit lines with unevenness.< >
An intelligent multisensor robotic system consisting of vision, tactile, and force/torque sensors is constructed. An artificial intelligence tool in the form of an expert system shell is included in the system to prov...
详细信息
An intelligent multisensor robotic system consisting of vision, tactile, and force/torque sensors is constructed. An artificial intelligence tool in the form of an expert system shell is included in the system to provide a high-level decision-making tool for the robotic system. The multisensor system consists of a VAX-11/750 supervisory computer, two PUMA-560 robots, a vision system, a work-cell tactile array sensor, an instrumented servo gripper equipped with force/torque and tactile array sensors, a six-component wrist force/torque sensor, and an expert system shell running on an IBM PC. All the components in the system are connected to the supervisory VAX computer. Software and hardware links of the system were also developed. Modularity was emphasized in the system development to permit easy and rapid reconfiguration of the system. A simple sensor fusion experiment was conducted. The experiment was used to test the software and hardware of the entire multisensor system and to illustrate the concept of sensor fusion.< >
暂无评论