Robotic TCF Calibration Based on a Planar Template
Abstract
For the TCF (tool/terminal control frame) calibration of a robot holding a LDS (laser displacement sensor), a method based on a planar template is proposed. The robot is required to operate its end-effector so that the LDS measures the planar template at different configurations. Then the calibration parameters are solved by nonlinear least square fitting. To reduce the singularity of the problem, the parameter-control strategy during the calibration is qualitatively analyzed. Except for a planar template with high surface accuracy, no extra measurement equipments are needed. The calibration process is simple and easy for operation and automatization. Simulation and experiment results show that the method presented in this paper has high accuracy.
Aiming at the calibration problem of the position parameter of camera and screen in vision-based specular surface measurement, we propose a fast and simple calibration method based on mirror. A LCD screen is used to project a new circle-based pattern, then a camera captures the image of the pattern in the mirror, and the camera calibration is performed. The internal and external parameters of the camera are obtained after camera calibration. We can obtain the real position of the calibration pattern according to the theory of mirror imaging. In the calibration process the mirror can be moved freely without knowing its positions. Simulation and real experiment results show that the proposed method is simple and fast, and has high calibration accuracy, which satisfies the requirement of actual measurement.
In order to calibrate the position and orientation of laser displacement sensors mounted on a robot, an approach was devised for calibrating the position and orientation parameters simultaneously, based on the model of plane and sphere fitting. A contrast experiment was proposed and conducted systematically based on robot off-line programming and robot kinematics simulation. Effects on the accuracy of the two calibration algorithm were also analyzed such as the measurement error, length of sensor and the number of calibration parameters. Observability indices were introduced to evaluate the efficiency of parameter identification with the measurement configurations in robot motion space. Optimal measurement configurations and best evaluation index were calculated using on optimization algorithm. The effects of the non-modeling error and measurement error are reduced. The stability, accuracy and efficiency of the algorithm are improved. Experiment results of our improved calibration algorithm indicate that the standard deviation of position is 0.37 mm, the standard deviation of orientation is 0.35°.
An accurate TCF (tool control frame) model is essential for high-accuracy robot off-line programming. Meanwhile, TCF calibration is an important procedure for production recovery after robot collides in industrial field. This article proposes a novel TCF calibration method in robotic visual measurement system in which the robot TCF is defined based on the model of visual sensor and a standard sphere with known diameter is utilized as the calibration target. With the translational and rotational movements of the industrial robot, the visual senor measures the center of standard sphere from multiple different robot postures, TCF orientation and TCP position are determined in two steps. Robot off-line programming is performed based on the TCF calibration result, and robot collision is simulated on an ABB IRB2400 industrial robot. Experimental results have validated the effectiveness and efficiency of the standard sphere-based TCF calibration method, which could control the deviation of an identical featured point within 0.5 mm measured before and after collision recovery.
A stretching device with intelligent control was developed to treat spastic/contractured ankles of neurologically impaired patients and evaluate the outcome. The device stretched the ankle joint safely throughout the range of motion (ROM) to extreme positions until a specified peak resistance torque was reached with the stretching velocity controlled based on the resistance torque. The ankle was then held at the extreme position for a period of time to let the stress relax before it was rotated back to the other extreme position. The stretching was slow at the joint extreme positions, making it possible to reach a larger ROM safely, and it was fast in the middle ROM so the majority of the treatment was spent in stretching the problematic extreme ROM. The device was evaluated in five healthy subjects and used to treat four stroke patients. Furthermore, it was used to evaluate the treatment outcome in multiple aspects, including passive/active joint ROM, stiffness, viscous damping, and reflex excitability. The intelligent control and yet simple design of the device suggest that the device can be made portable at relatively low cost, making it available to patients/therapists for frequent use in clinics/home and allowing more effective treatment and long-term improvement.
Zhuang et al. (1994) proposed a method that allows simultaneous
computation of the rigid transformations from world frame to robot base
frame and from hand frame to camera frame. Their method attempts to
solve a homogeneous matrix equation of the for, AX=ZB. They use
quaternions to derive explicit linear solution for X and Z. In this
paper, we present two new solutions that attempt to solve the
homogeneous matrix equation mentioned above: 1) a closed-form method
which uses quaternion algebra and a positive quadratic error function
associated with this representation; 2) a method based on nonlinear
constrained minimization and which simultaneously solves for rotations
and translations. These results may be useful to other problems that can
be formulated in the same mathematical form. We perform a sensitivity
analysis for both our two methods and the linear method developed by
Zhuang et al. This analysis allows the comparison of the three methods.
In the light of this comparison, the nonlinear optimization method,
which solves for rotations and translations simultaneously, seems to be
the most stable one with respect to noise and to measurement
errors
We present a new hand-eye calibration method on the basis of controlling motion of manipulator mounted with a camera. Different from other algorithms, this method utilizes nominal rotation for camera coordinate to make rotation transform into translation. The algorithm needs the manipulator to execute two translation motions and a rotation motion, and requires only two feature points in the scene. The algorithm is both convenient and practical in application. At the same time we also present an active-vision-based calculational method for depth value of points in the scene.
With the application and development of off-line programming system (OLPS), as one of the key techniques for the application of OLP technology, the robot calibration technique is drawing more and more attention. Aiming at part calibration for the application of robotic off-line programming, the part calibration algorithm is further discussed for the adjustment and matching of part models by transformation matrix between part coordinates in the simulation environments and real workspace. Three part calibration methodologies, calibration of orthogonal plane part, four-point calibration of circular reference part, and three-point calibration of auxiliary feature points, are presented. Part calibration can be done by using the algorithm proposed only.
A robot can be used for noncontact measurements when a laser displacement sensor is fixed to its end-link. The robot's end-link and the laser displacement sensor can be calibrated by measuring a fixed point from different poses (FPDP) method. In this method, one point is measured from various robot's poses multiple times. The homogeneous transformation matrix between the coordinates of the end-link and the laser displacement sensor can be quickly calculated using the equations deduced from the robot kinematical equations and the least-squares principle. Experimental measurements of the depth of a pit show that the method is effective and quick. Further work will apply this method to the detection of ablation in hydroturbine blades in the Three Gorges Dam.
A self-calibration approach to hand-eye relation of robot based on a single point in the scene is proposed. The manipulator end-effector is accurately controlled to perform five (or more) pure translational motions and two (or more) pure rotational motions, and the camera is required to image a single point in the scene. Motions of the camera are estimated based on the disparity and depth value of the point, relative position constraint equations between coordinates of the manipulator end-effector and the camera are set up, and the intrinsic parameters of camera and the hand-eye relation are obtained linearly. In the calibration process, only one point in the scene needs to be extracted, and neither matching nor orthogonal motion is required, therefore, motions of the manipulator can be controlled conveniently and the algorithm can be implemented simply. Experiment results of both simulation data and real image data show that the proposed method is effective and feasible.
For off-line programming to work, systematic methods must be developed to account for non-ideal performance of the parts and devices in the manufacturing cell. Although much of the literature focuses on robot inaccuracy, this paper considers practical methods for the tool control frame (TCF) calibration and rigid-body compensation required to close the inverse kinematics loop for target driven tasks.
In contrast to contemporary estimation methods, a closed-form, easily automated, solution is introduced for calibrating the position and orientation (pose) of orthogonal end-effectors when the distal robot joint is revolute. This paper also considers methods for measuring and compensating the small rigid-body perturbations that result from non-repeatable part delivery systems or from geometric distortion. These methods are designed to eliminate r θ error from the rigid-body prediction and can be conducted in real-time. Without accurate TCF calibration and rigid-body compensation, even the most accurate robot will fail to complete an off-line programmed task if the task tolerances are stringent.
Haptic based human–computer interaction (HapHCI) system is currently the frontier of robot research, which is widely used in virtual reality, rehabilitation, entertainment, and so on. The measurement of the multi-dimensional interactive force between human hand and interaction device such as hand-controller, joystick, limb rehabilitation device, etc., becomes an important component of the HapHCI. However, the existing commercial 6 degree-of-freedom (DOF) force sensors are too expensive and often over designed for HapHCI not only in axis but also in bandwidth. In this paper, a novel 4 DOF wrist force/torque sensor suitable for HapHCI is developed, which is self-decoupled without calculating the decoupling matrix. Thus this type of wrist force/torque can be called as direct output force sensor, which is quite different from the conventional ones called as indirect output force sensor. Its elastic body has a simple geometry which is easy to fabricate, and half the number of strain gauges compared to the existing commercial 6 DOF force/torque sensor. So the 4 DOF force/torque sensor is much lower in cost. This paper introduces the elastic body structure of the wrist force/torque sensors, and analyses the self-decoupled principle in detail. A prototype sensor is fabricated, and the calibration test results of the 4 DOF force/torque sensor validate the analysis and demonstrate the advantage of such a sensor.
An intelligent stretching device was developed to treat the spastic/contractured ankle of neurologically impaired patients. The device stretched the ankle safely throughout the range of motion (ROM) to extreme dorsiflexion and plantarflexion until a specified peak resistance torque was reached with the stretching velocity controlled based on the resistance torque. The ankle was held at the extreme position for a period of time to let stress relaxation occur before it was rotated back to the other extreme position. Stretching was slow at the joint extreme positions, making it possible to reach a larger ROM safely and it was fast in the middle ROM so the majority of the treatment was spent in stretching the problematic extreme ROM. Furthermore, the device evaluated treatment outcome quantitatively in multiple aspects, including active and passive ROM, joint stiffness and viscous damping and reflex excitability. The stretching resulted in considerable changes in joint passive ROM, stiffness, viscous damping and reflex gain. The intelligent control and yet simple design of the device suggest that with appropriate simplification, the device can be made portable and low cost, making it available to patients and therapists for frequent use in clinics/home and allowing more effective treatment and long-term improvement.