In this paper, we present a novel continuous image registration method (CIRM), which yields near-zero bias and has high computational efficiency. It can be realized for real-time position estimation to enable ultraprecise 2-D motion tracking and motion control over a large motion range. As the two variables of the method are continuous in spatial domain, pixel-level image registration is unnecessary, thus the CIRM can continuously track the moving target according to the incoming target image. When applied to a specific target object, measurement resolution of the method is predicted according to the reference image model of the object along with the variance of the camera's overall image noise. The maximum permissible target speed is proportional to the permissible frame rate, which is limited by the required computational time. The precision, measurement resolution, and computational efficiency of the method are verified through computer simulations and experiments. Specifically, the CIRM is implemented and integrated with a visual sensing system. Near-zero bias, measurement resolution of 0.1 nm (0.0008 pixels), and measurement of one nanometer stepping are demonstrated.

This paper presents the development of a visual servo control system that is capable of trapping and steering a microscopic magnetic bead in a 3-D workspace using a hexapole magnetic actuator. The magnetic actuator employs six sharp-tipped magnetic poles, arranged in a hexagonal configuration, and six actuating coils to achieve 3-D magnetic actuation. Real-time 3-D visual tracking is realized through processing video images acquired by a complementary metal-oxide semiconductor (CMOS) camera to achieve nanometer-scale resolution for particle tracking. A model-based nonlinear control law is developed to achieve two objectives: 1) to stabilize the magnetically propelled microscopic bead and control its Brownian motion and 2) to rapidly steer the microscopic bead within the 3-D workspace. Specifically, 3-D visual servo control of a 2.8-μm magnetic bead is employed to validate the capability of the developed control system. The control performances in terms of variance minimization, 100 nm stepping, and large range steering are experimentally demonstrated.

This paper presents the theoretical analysis and experimental investigation of actively controlled manipulation of a magnetic microbead using quadrupole magnetic tweezers. Bead dynamics, magnetic actuation, and visual measurement are analyzed. A feedback control law is developed and implemented to stabilize and steer the motion of the magnetic microbead. It is developed in two steps. First, an inverse model, which is associated with a lumped-parameter analytical force model, is derived to enable feedback linearization. Second, linear controllers are designed to achieve motion stabilization and manipulation of the magnetic microbead. A proportional-gain controller along with feedback linearization is implemented to establish a stable trapping of the magnetic bead to facilitate system calibration. Experiments are then performed to validate the derived inverse force model and theoretical analysis. In addition, a minimum-variance controller is designed and employed to reduce the variance of the bead's Brownian motion. The control performance in terms of variance reduction, nanostepping, and large-range steering is then experimentally demonstrated.

This paper presents a three-axis vision motion sensor and its applications to visual servo control. The vision sensor is integrated with a three-axis piezo stage to form a visual servo control system that achieves nanometer resolution in all threex-y-zmotion axes. Motion measurement is achieved using a single interferometer-equipped optical microscope. A real-time image-processing algorithm that processes interference fringe patterns and that achieves nanometer out-of-plane resolution is presented. Furthermore, a feedback-control scheme is introduced to control the sensor plane using an Objective-Z-Positioner to enable automatic tracking of moving objects. It expands the out-of-plane measurement range of the vision sensor beyond its inherent depth of field of several micrometers to 100 mum and beyond. An integrated visual servo system is implemented and experimental results are shown.

This paper presents the development of a compact six-axis magnetic levitation stage, including the design and implementation of magnetic actuators, laser interferometer motion sensors, and motion controllers. The designed travel volume of the stage is 2times2times2 mm in translation and 4degtimes4degtimes4deg in rotation. A two-axis linear actuator, based on the magnetic Lorentz force law, was designed, and three two-axis actuators, equivalent to six single -axis actuators, were implemented to achieve six-axis actuation. A high-resolution laser interferometer measurement system was implemented and employed to measure the six-axis motion of the stage, facilitating real-time feedback control. Feedback linearization, based on rigid-body dynamics of the levitated stage, and force distribution were implemented in a computer-controlled architecture so as to establish a decoupled dynamics between the six computed inputs and the resulting six-axis motions. Constant gain controllers were then designed and implemented, according to the concept of loop shaping, for each of the six axes, and high positioning stability, 1.1 nm root mean square (RMS) for x and 0.74 nm RMS for y, has been achieved. Experimental results are presented to illustrate positioning stability, system linearity, dynamic response invariance, nano stepping, multiaxis contouring, and large rotational motion

This paper presents the development of a novel active visual measurement technique, laterally sampled white-light interferometry (L-SWLI), which is capable of real-time visual tracking of six-degree-of-freedom (6-DOF) rigid body motion with near-nanometer precision. The visual tracking system is integrated with a 6-DOF motion stage to realize an ultraprecision six-axis visual servo-control system. Contrary to conventional interferometric techniques, L-SWLI obtains the complete pose of the target object from a single image frame, therefore allowing real-time tracking. Six-DOF motions are obtained from measuring the fringe pattern on multiple surfaces of the object or from a single surface with additional information gained from conventional image-processing techniques. The feasibility of the visually servoed motion scheme was demonstrated on a micro cantilever. The cantilever was maneuvered in a three-dimensional space with near-nanometer motion resolution in all three translational axes.

A systematic approach is proposed to automatically extract geometric surface features from a point cloud composed of a set of unorganized three-dimensional coordinate points by data segmentation. The point cloud is sampled from the boundary surface of a mechanical component with arbitrary shape. The proposed approach is composed of three steps. In the first step, a mesh surface domain is reconstructed to establish an explicit topological relation among the discrete points. The topological adjacency is further optimized to recover the second order object geometry. In the second step, curvature-based border detection is applied on the irregular mesh to extract both sharp borders with tangent discontinuity and smooth borders with curvature discontinuity. Finally, the mesh patches separated by the extracted borders are grouped together in the third step. For objects with complex shape, a multilevel segmentation scheme is proposed for better results. The capability of the proposed approach is demonstrated using various point clouds having distinct characteristics. Integrated with state of art scanning devices, the developed segmentation scheme can support reverse engineering of high precision mechanical components. It has potential applications in a whole spectrum of engineering problems with a major impact on rapid design and prototyping, shape analysis, and virtual reality.

An automatic camera calibration scheme that utilizes a coordinate measuring machine (CMM) and a camera calibration algorithm is presented for a multiple-sensor integrated coordinate measurement system. In the proposed calibration scheme, the touch probe tip carried by the CMM is employed to automatically generate high-precision calibration target points for camera calibration and sensor integration. A camera calibration algorithm with analytical formulations is developed to calibrate camera parameters in three stages without nonlinear minimization procedures. Simulations and experiments were performed to verify the proposed camera calibration algorithm. The precision of the automatic camera calibration scheme is also evaluated.

The authors present an optimal match scheme that aligns the measurement data with the design data in CAD-directed dimensional inspection. The principle of this scheme is to determine the actual measured points on the nominal surface by minimizing the sum of the squared distances of the measurement data from the surface with respect to the parameters of a rigid body transformation. Its computational efficiency and robustness against singularity are discussed. Two applications are examined. The first is determining the design coordinate system with respect to a machine coordinate system. The second is comparative analysis for precision measurement. For determining the design coordinate system, a sensitivity measure is proposed and confirmed experimentally. For error comparative analysis, a statistical model is developed to determine the minimum number of required measurement points so that the sampled points can closely represent the entire population.>

A hierarchical planning system is presented which uses heuristics for path planning in dimensional inspection using coordinate measuring machines (CMMs). The objective is to automate the planning of a collision-free inspection path. The system demonstrates that the inspection paths of geometrically complex real world parts with multiple features can automatically be generated with efficiency. This is made possible by appropriate CMM abstractions, genuine 3-D collision detection, and heuristic modifications of reasonably assumed default inspection paths. Automatic selection and modification of probe angle based on a local accessibility analysis of a given feature are also presented. The collision-free inspection path can then be simulated in a CAD environment before it is carried out by the real CMM.>

Experiments on the payload adaptation of a flexible one-link manipulator with an unknown payload on the tip are described. Off line and online system identification schemes are implemented and compared. Three robust online indentification schemes, namely, the standard recursive-least-squares method, a least-squares method with known information lumped, and an average modified gradient method, are implemented, and their performance is evaluated based on the models obtained from off line identification. Through the use of an eight-order Butterworth low-pass filter, the effect of unmodeled dynamics on the online identification schemes is investigated. It is found, through experiments, that the least-squares-based methods are sensitive to unmodeled dynamics, while the average gradient method shows significant improvement in counteracting the unmodeled dynamics.>

The selection of measurement configurations for robot calibration is investigated. The goal is to select a set of robot measurement configurations which will yield maximum observability of the model parameter errors so that the effects of noise in parameter estimation can be minimized. The noise considered includes both measurement and modeling errors. An observability measure is used as a criterion for selecting measurement configurations in calibration. Experimental studies are performed to demonstrate the importance of observability to parameter estimation and to verify its implications in robot calibration. Based on the observability measure defined, the optimal measurement configurations for kinematic model calibration are determined for PUMA-type robots.>

For the accurate control and implementation of robots in an integrated manufacturing environment, a knowledge of the position errors of the end effector is important. The authors present a framework which can be used to characterize the position errors of robot manipulators. The objectives are to study the statistical properties of the errors and to characterize their distribution over the work space. The implications of the position-error field of a robot in its design and applications are also discussed.>

The modeling and control of a lightweight flexible manipulator with end-point sensing/feedback are studied. Using the FEM (finite-element method) scheme in modeling a flexible manipulator, a lower-order compensator can be accomplished. The performance of this near-optimal compensator has been investigated through simulations. The simulation results have shown that the implementation of a payload-adaptation synthesizer, which synthesizes a payload identified and a nominal regulator/estimator interpolator to obtain a near-optimal compensator, has been successful in identifying the payload which is applied on the tip. The synthesizer provides a near-optimal damping for this noncolocated control system with torque constraint on the actuator.>

A theoretical study on the dexterous workspace of robotic manipulators is presented. For a robot with wrists that can generate a full range of orientations, the boundary of the robot's dexterous workspace is governed by the boundary of W/sub 1/(4), where W/sub 1/(4) is the reachable space of joint 4 when joints 1-3 are free to rotate. A method is developed on this basis. Three examples are given to illustrate this concept and method. They show that for simple robots, analytical expressions for the dexterous workspace can be obtained.>