Uncalibrated Vision Research Articles

Integrating a Pipette Into a Robot Manipulator With Uncalibrated Vision and TCP for Liquid Handling

Integrating a Pipette Into a Robot Manipulator With Uncalibrated Vision and TCP for Liquid Handling

Breast Morphology Recognition and Control Based on Uncalibrated Vision

Breast Morphology Recognition and Control Based on Uncalibrated Vision

Trajectory tracking with ANFIS and uncalibrated vision system for uncertain robotic manipulators

This paper describes a new approach for tracking the trajectory of uncertain robotic manipulators using ANFIS (Artificial Neuro-Fuzzy Inference System) and uncalibrated vision system. The main emphasis of this work is on the ability to estimate the positioning accuracy and repeatability of a low-cost robotic arm with unknown parameters under uncalibrated vision system. In this study, captured image data are collected from two fixed-cameras vision system; installed on the top and lateral sides of the robot, respectively. For training and validating purposes, the robot is manoeuvring within its workspace using forward kinematics. The tracking system is trained using ANFIS with subtractive clustering method in MATLAB. Extensive simulations were carried out to illustrate the effectiveness of the proposed visual tracking method for LabVolt R5150 manipulator in our laboratory. Observing the simulation results, the performance of the proposed approach is efficient for using the vision-based learning system as visual feedback in uncertain robotic manipulator.

ANFIS-Based Visual Pose Estimation of Uncertain Robotic Arm Using Two Uncalibrated Cameras

This paper describes a new approach for the visual pose estimation of an uncertain robotic manipulator using ANFIS (Artificial Neuro-Fuzzy Inference System) and two uncalibrated cameras. The main emphasis of this work is on the ability to estimate the positioning accuracy and repeatability of a low-cost robotic arm with unknown parameters under uncalibrated vision system. The vision system is composed of two cameras; installed on the top and on the lateral side of the robot, respectively. These two cameras need no calibration; thus, they can be installed in any position and orientation with just the condition that the end-effector of the robot must remain always visible. A red-colored feature point is fixed on the end of the third robotic arm link. In this study, captured image data via two fixed-cameras vision system are used as the sensor feedback for the position tracking of an uncertain robotic arm. LabVolt R5150 manipulator in our laboratory is used as case study. The visual estimation system is trained using ANFIS with subtractive clustering method in MATLAB. In MATLAB, the robot, feature point and cameras are simulated as physical behaviors. To get the required data for ANFIS, the manipulator was maneuvered within its workspace using forward kinematics and the feature point image coordinates were acquired with the two cameras. Simulation experiments show that the location of the robotic arm can be trained in ANFIS using two uncalibrated cameras; and problems for computational complexity and calibration requirement of multi-view geometry can be eliminated. Observing Mean Square Error (MSE), Root Mean Square Error (RMSE), Error Mean and Standard Deviation Errors, the performance of the proposed approach is efficient for using as visual feedback in uncertain robotic manipulator. Further, the proposed approach using ANFIS and uncalibrated vision system has better in flexibility, user-friendly manner and computational concepts over conventional techniques.

Kalman Filter-Based Robot Manipulator Five-Degrees of Freedom Uncalibrated Vision Positioning

Robot manipulator position and posture control is a popular topic in the field of uncalibrated visual servoing, this paper presents a kalman filter-based robot manipulator five-degrees of freedom uncalibrated vision positioning method. In the case of the fixed binocular cameras and manipulator parameters are unknown; firstly, the specific point and angle image features information in the camera image space were selected in order to describe the relative pose relationship between robot manipulator ends and goals. Then, the kalman filter online estimation algorithm was applied to calculate image Jacobian matrix which is mapping relationship between image space to cartesian mission space, and vision controller was designed in the image plane realized robot manipulator five-degrees of freedom uncalibrated vision positioning control. Finally, Six-degrees of freedom robot manipulator’s five-degrees of freedom uncalibrated visual positioning Simulink model established in the Matlab environment, and the simulation result show that kalman filter online estimation method made the robot manipulator rapid convergence to the desired position and posture with high accuracy.

Path-Tracking Maneuvers With Industrial Robot Manipulators Using Uncalibrated Vision and Impedance Control

This paper presents an interaction control strategy for industrial robot manipulators which consists of the combination of a calibration-free, vision-based control method with an impedance-control approach. The vision-based, robot control method known as camera-space manipulation is used to generate a given, previously defined trajectory over an arbitrary surface. Then, a kinematic impedance controller is implemented in order to regulate the interaction forces generated by the contact between the robot end-effector and the work surface where the trajectory is traced. The paper presents experimental evidence on how the vision-force sensory fusion is applied to a path-tracking task, using a Fanuc M16-iB industrial robot equipped with a force/torque sensor at the wrist. In this task, several levels of interaction force between the robot end-effector and the surface were defined. As discussed in the paper, such a synergy between the control schemes is seen as a viable alternative for performing industrial maneuvers that require force modulation between the tool held by the robot and the working surface.

Enhanced Assessment of the Wound-Healing Process by Accurate Multiview Tissue Classification

With the widespread use of digital cameras, freehand wound imaging has become common practice in clinical settings. There is however still a demand for a practical tool for accurate wound healing assessment, combining dimensional measurements and tissue classification in a single user-friendly system. We achieved the first part of this objective by computing a 3-D model for wound measurements using uncalibrated vision techniques. We focus here on tissue classification from color and texture region descriptors computed after unsupervised segmentation. Due to perspective distortions, uncontrolled lighting conditions and view points, wound assessments vary significantly between patient examinations. The main contribution of this paper is to overcome this drawback with a multiview strategy for tissue classification, relying on a 3-D model onto which tissue labels are mapped and classification results merged. The experimental classification tests demonstrate that enhanced repeatability and robustness are obtained and that metric assessment is achieved through real area and volume measurements and wound outline extraction. This innovative tool is intended for use not only in therapeutic follow-up in hospitals but also for telemedicine purposes and clinical research, where repeatability and accuracy of wound assessment are critical.

Force-Contouring Process Using Uncalibrated Vision and Impedance Control

This paper presents an interaction control strategy for industrial robot manipulators which consists of the combination of a calibration-free, vision-based control method with an impedance control approach. The vision-based, robot control method known as camera-space manipulation is used to map a given, previously-defined trajectory over an arbitrary surface. Then, a kinematic posture-based impedance controller is implemented in order to regulate the interaction forces and torques generated by the contact between the robot end-effector and the work surface where the trajectory is defined. The paper presents experimental evidence on how the vision-force sensory fusion is useful to implement a force contouring task in a machining process, by using a Fanuc M16-iB industrial robot equipped with a wrist force/torque sensor.

An optimal path-generation algorithm for manufacturing of arbitrarily curved surfaces using uncalibrated vision

Multiple industrial manufacturing tasks require a complex path to be followed precisely over an arbitrary surface which has a geometry that is not known with precision. Examples of such tasks include welding, glue-application, cutting, plasma-spraying, etc., over commercial plates whose geometry cannot be known in advance. Such processes are in general referred to as surface manufacturing. In this work, a path is traced over the surface in an optimal fashion, using the concept of geodesic mapping. By definition, a geodesic line is the shortest line that joins two points over a surface whose algebraic representation is known. Such an optimal solution of a problem, associated with variational calculus, is the approach employed for mapping complex paths, defined in a data base, over a surface of arbitrary geometry. The straight-line segments in which a complex path can be divided are mapped onto an arbitrary surface as geodesic lines. The presented algorithm enables a user to interact with the system in a simple and efficient manner using a commercial computer pointing device. The algorithm was tested experimentally in an industrial maneuver involving arc welding, using an industrial robot and a method of vision-based, robot guidance known as camera-space manipulation. This method has the advantage of not requiring calibration of optical or mechanical components.

An efficient multi-camera, multi-target scheme for the three-dimensional control of robots using uncalibrated vision

A vision-based control methodology is presented in this paper that can perform accurate, three-dimensional (3D), positioning and path-tracking tasks. Tested with the challenging manufacturing task of welding in an unstructured environment, the proposed methodology has proven to be highly reliable, consistently achieving terminal precision of 1 mm . A key limiting factor for this high precision is camera–space resolution per unit physical space. This paper also presents a means of preserving and even increasing this ratio over a large region of the robot's workspace by using data from multiple vision sensors. In the experiments reported in this paper, a laser is used to facilitate the image processing aspect of the vision-based control strategy. The laser projects “laser spots” over the workpiece in order to gather information about the workpiece geometry. Previous applications of the control method were limited to considering only local, geometric information of the workpiece, close to the region where the robot's tool is going to be placed. This paper presents a methodology to consider all available information about the geometry of the workpiece. This data is represented in a compact matrix format that is used within the algorithm to evaluate an optimal robot configuration. The proposed strategy processes and stores the information that comes from various vision sensors in an efficient manner. An important goal of the proposed methodology is to facilitate the use of industrial robots in unstructured environments. A graphical-user-interface (GUI) has been developed that simplifies the use of the robot/vision system. With this GUI, complex tasks such as welding can be successfully performed by users with limited experience in the control of robots and welding techniques.

Uncalibrated vision based on lines for robot navigation

In this paper, we study an uncalibrated monocular vision system based on lines for robot navigation in man-made environments. Neither environment a priori map is used nor euclidean reconstruction obtained. The method is based on the extracted straight lines, supposing the robot moves on a plane ground. The effect in the image of camera rotation is computed from the homography of the line at infinity. From corresponding vertical lines in two uncalibrated images both, the robot heading and a region in the image, which correspond to the free space ahead, are computed. This is the most relevant aspect of our method, because usually it has been considered that corresponding lines in two images do not constraint the structure and motion problem. However, using the visibility constraint and compensating the camera rotation, we have obtained the direction of translation and a qualitative free space ahead. A control scheme in image coordinates has been implemented to carry out the experiments with our robot moving in the lab. It allows the robot to advance along a corridor avoiding doors or pillars without collision. The computation of information from lines is accurate enough and although the robot has not previous map, it is autonomous when moving in this environment. The method supposes advantages in comparison with other works because neither image nor camera-wheel calibration are required.

視覚サーボと内力の制御の統合に基づく多指ハンドによる操り

In this paper, we describe a control method for a multi-fingered hand with uncalibrated vision to manipulate an object. To achieve a robust and quick manipulation without precise calibration, adaptive visual servoing is applied. on the other hand, internal force control also has to be applied to maintain grasping. To integrate these controllers, we utilize the concepts of visual impedance and the principle of virtual work, and eliminate the interference. Experimental results are shown to demonstrate the validity of the proposed scheme.

Computational Vision at Yale

We present a brief introduction to the five articles that make up this special issue: Shock Graphs and Shape Matching, The Bas-Relief Ambiguity, Incremental Focus of Attention for Robust Vision Based Tracking, What Tasks can be Performed with an Uncalibrated Stereo Vision System, and Volumetric Deformation Analysis Using Mechanics-Based Data Fusion: Application in Cardiac Motion Recovery. Tjis introduction and accompanying articles provide a by no means exhaustive, but hopefully representative sampling of the computational vision at Yale University.

Automatic feature matching in uncalibrated stereo vision through the use of color

One difficult problem in stereo vision is how to establish correspondence between features extracted from a pair of images. The difficulty is due to ambiguities or inconsistencies of available information on images. In this paper, we invetigate stereo correspondence problem in the framework of color stereo vision. We propose the use of a matching consistence (MC) constraint in RGB color space and the generalized epipolar geometry to develop an automatic feature matching algorithm.