Robots--Control systems

This research introduces a unified approach to visual looming. Visual looming is related to an increasing projected size of an object on a viewer's retina while the relative distance between the viewer and the object decreases. Psychophysicists and neurobiologists have studied this phenomenon by observing vision and action in unison and have reported subject's tendency to react defensively or using this information in an anticipatory control of the body. Since visual looming induces senses of threat of collision, the same cue, if quantified, can be used along with visual fixation in obstacle avoidance in mobile robots. In quantitative form visual looming is defined as the time derivative of the relative distance (range) between the observer and the object divided by the relative distance itself. The visual looming is a measurable variable. Following the paradigm of Active Vision the approach in this research uses visual fixation to selectively attend a small part of the image, that is relevant to the task. Visual looming provides a time-based mapping from a "set of 2-D image cues" to "time-based 3-D space". This research describes how visual looming, which is a concept related to an object in the 3-D world, can be calculated studying the relative temporal change in the following four different attributes of a sequence of 2-D images: (i) image area; (ii) image brightness; (iii) texture density in the image; (iv) image blur. From a simple closed form expression it shows that a powerful unified approach can be adopted in these methods. An extension of this unified approach establishes a strong relationship with the Weber-Fechner law in Psychophysics. The four different methods explored for the calculation of looming are simple. The experimental results illustrate how the measured values of looming stay close to the actual values. This research also introduces one important visual invariant $\Re$ that exists in relative movements between a camera light-source pair and a visible object. Finally, looming is used in the sense of a threat of collision, to navigate in an unknown environment. The results show that the approach can be used in real-time obstacle avoidance with very little a-priori knowledge.

Parallel manipulators have their special characteristics in contrast to the traditional serial type of robots. Stewart platform is a typical six degree of freedom fully parallel robot manipulator. The goal of this research is to enhance the accuracy and the restricted workspace of the Stewart platform. The first part of the dissertation discusses the effect of three kinematic constraints: link length limitation, joint angle limitation and link interference, and kinematic parameters on the workspace of the platform. An algorithm considering the above constraints for the determination of the volume and the envelop of Stewart platform workspace is developed. The workspace volume is used as a criterion to evaluate the effects of the platform dimensions and kinematic constraints on the workspace and the dexterity of the Stewart platform. The analysis and algorithm can be used as a design tool to select dimensions, actuators and joints in order to maximize the workspace. The remaining parts of the dissertation focus on the accuracy enhancement. Manufacturing tolerances, installation errors and link offsets cause deviations with respect to the nominal parameters of the platform. As a result, if nominal parameters are being used, the resulting platform pose will be inaccurate. An accurate kinematic model of Stewart platform which accommodates all manufacturing and installation errors is developed. In order to evaluate the effects of the above factors on the accuracy, algorithms for the forward and inverse kinematics solutions of the accurate model are developed. The effects of different manufacturing tolerances and installation errors on the platform accuracy are investigated based on this model. Simulation results provide insight into the expected accuracy and indicate the major factors contributing to the inaccuracies. In order to enhance the accuracy, there is a need to calibrate the platform, or to determine the actual values of the kinematic parameters (Parameter Identification) and to incorporate these into the inverse kinematic solution (Accuracy Compensation). An error-model based algorithm for the parameter identification is developed. Procedures for the formulation of the identification Jacobian and for accuracy compensation are presented. The algorithms are tested using simulated measurements in which the realistic measurement noise is included. As a result, pose error of the platform are significantly reduced.

The dissertation focuses on robot manipulator dynamic modeling, and inertial
and kinematic parameters identification problem. An automatic dynamic parameters
derivation symbolic algorithm is presented. This algorithm provides the linearly
independent dynamic parameters set. It is shown that all the dynamic parameters are
identifiable when the trajectory is persistently exciting. The parameters set satisfies
the necessary condition of finding a persistently exciting trajectory. Since in practice the system data matrix is corrupted with noise, conventional
estimation methods do not converge to the true values. An error bound is given for
Kalman filters. Total least squares method is introduced to obtain unbiased
estimates.
Simulations studies are presented for five particular identification methods.
The simulations are performed under different noise levels.
Observability problems for the inertial and kinematic parameters are
investigated. U%wer certain conditions all L%wearly Independent Parameters
derived from are observable.
The inertial and kinematic parameters can be categorized into three parts
according to their influences on the system dynamics. The dissertation gives an
algorithm to classify these parameters.