Use of neural networks in robot positioning of large flexible redundant manipulators

Deflection compensation of flexible boom structures in robot positioning is usually done using tables containing the magnitude of the deflection with inverse kinematics solutions of a rigid structure. The number of table values increases greatly if the working area of the boom is large and the required positioning accuracy is high. The inverse kinematics problems are very nonlinear, and if the structure is redundant, in some cases it cannot be solved in a closed form. If the structural flexibility of the manipulator arms is taken into account, the problem is almost impossible to solve using analytical methods. Neural networks offer a possibility to approximate any linear or nonlinear function. This study presents four different methods of using neural networks in the static deflection compensation and inverse kinematics solution of a flexible hydraulically driven manipulator. The training information required for training neural networks is obtained by employing a simulation model that includes elasticity characteristics. The functionality of the presented methods is tested based on the simulated and measured results of positioning accuracy. The simulated positioning accuracy is tested in 25 separate coordinate points. For each point, the positioning is tested with five different mass loads. The mean positioning error of a manipulator decreased from 31.9 mm to 4.1 mm in the test points. This accuracy enables the use of flexible manipulators in the positioning of larger objects. The measured positioning accuracy is tested in 9 separate points using three different mass loads. The mean positioning error decreased from 10.6 mm to 4.7 mm and the maximum error from 27.5 mm to 11.0 mm.

The Sottunga-Jurmo shear zone : structure and deformation history of a crustal-scale ductile shear zone in SW Finland

The aim of this study is to gain a better understanding of the structure and the deformation history of a NW-SE trending regional, crustal-scale shear structure in the Åland archipelago, SW Finland, called the Sottunga-Jurmo shear zone (SJSZ). Approaches involving e.g. structural geology, geochronology, geochemistry and metamorphic petrology were utilised in order to reconstruct the overall deformation history of the study area. The study therefore describes several features of the shear zone including structures, kinematics and lithologies within the study area, the ages of the different deformation phases (ductile to brittle) within the shear zone, as well as some geothermobarometric results. The results indicate that the SJSZ outlines a major crustal discontinuity between the extensively migmatized rocks NE of the shear zone and the unmigmatised, amphibolite facies rocks SW of the zone. The main SJSZ shows overall dextral lateral kinematics with a SW-side up vertical component and deformation partitioning into pure shear and simple shear dominated deformation styles that was intensified toward later stages of the deformation history. The deformation partitioning resulted in complex folding and refolding against the SW margin of the SJSZ, including conical and sheath folds, and in a formation of several minor strike-slip shear zones both parallel and conjugate to the main SJSZ in order to accommodate the regional transpressive stresses. Different deformation phases within the study area were dated by SIMS (zircon U-Pb), ID-TIMS (titanite U-Pb) and 40Ar/39Ar (pseudotachylyte wholerock) methods. The first deformation phase within the ca. 1.88 Ga rocks of the study area is dated at ca. 1.85 Ga, and the shear zone was reactivated twice within the ductile regime (at ca. 1.83 Ga and 1.79 Ga), during which the strain was successively increasingly partitioned into the main SJSZ and the minor shear zones. The age determinations suggest that the orogenic processes within the study area did not occur in a temporal continuum; instead, the metamorphic zircon rims and titanites show distinct, 10-20 Ma long breaks in deformation between phases of active deformation. The results of this study further imply slow cooling of the rocks through 600-700ºC so that at 1.79 Ga, 2 the temperature was still at least 600ºC. The highest recorded metamorphic pressures are 6.4-7.1 kbar. At the late stages or soon after the last ductile phase (ca. 1.79 Ga), relatively high-T mylonites and ultramylonites were formed, witnessing extreme deformation partitioning and high strain rates. After the rocks reached lower amphibolite facies to amphibolite-greenschist facies transitional conditions (ca. 500-550ºC), they cooled rapidly, probably due to crustal uplift and exhumation. The shear zone was reactivated at least once within the semi-brittle to brittle regime between ca. 1.79 Ga and 1.58 Ga, as evidenced by cataclasites and pseudotachylytes. In summary, the results of this study suggest that the Sottunga-Jurmo shear zone (and the South Finland shear zone) defines a major crustal discontinuity, and played a central role in accommodating the regional stresses during and after the Svecofennian orogeny.

Multifrequency VLBI Observations of Selected Active Galactic Nuclei

In this dissertation, active galactic nuclei (AGN) are discussed, as they are seen with the high-resolution radio-astronomical technique called Very Long Baseline Interferometry (VLBI). This observational technique provides very high angular resolution (_ 10−300 = 1 milliarcsecond). VLBI observations, performed at different radio frequencies (multi-frequency VLBI), allow to penetrate deep into the core of an AGN to reveal an otherwise obscured inner part of the jet and the vicinity of the AGN’s central engine. Multi-frequency VLBI data are used to scrutinize the structure and evolution of the jet, as well as the distribution of the polarized emission. These data can help to derive the properties of the plasma and the magnetic field, and to provide constraints to the jet composition and the parameters of emission mechanisms. Also VLBI data can be used for testing the possible physical processes in the jet by comparing observational results with results of numerical simulations. The work presented in this thesis contributes to different aspects of AGN physics studies, as well as to the methodology of VLBI data reduction. In particular, Paper I reports evidence of optical and radio emission of AGN coming from the same region in the inner jet. This result was obtained via simultaneous observations of linear polarization in the optical and in radio using VLBI technique of a sample of AGN. Papers II and III describe, in detail, the jet kinematics of the blazar 0716+714, based on multi-frequency data, and reveal a peculiar kinematic pattern: plasma in the inner jet appears to move substantially faster that that in the large-scale jet. This peculiarity is explained by the jet bending, in Paper III. Also, Paper III presents a test of the new imaging technique for VLBI data, the Generalized Maximum Entropy Method (GMEM), with the observed (not simulated) data and compares its results with the conventional imaging. Papers IV and V report the results of observations of the circularly polarized (CP) emission in AGN at small spatial scales. In particular, Paper IV presents values of the core CP for 41 AGN at 15, 22 and 43 GHz, obtained with the help of the standard Gain transfer (GT) method, which was previously developed by D. Homan and J.Wardle for the calibration of multi-source VLBI observations. This method was developed for long multi-source observations, when many AGN are observed in a single VLBI run. In contrast, in Paper V, an attempt is made to apply the GT method to single-source VLBI observations. In such observations, the object list would include only a few sources: a target source and two or three calibrators, and it lasts much shorter than the multi-source experiment. For the CP calibration of a single-source observation, it is necessary to have a source with zero or known CP as one of the calibrators. If the archival observations included such a source to the list of calibrators, the GT could also be used for the archival data, increasing a list of known AGN with the CP at small spatial scale. Paper V contains also calculation of contributions of different sourced of errors to the uncertainty of the final result, and presents the first results for the blazar 0716+714.

Tactile-Proprioceptive Robotic Grasping

Robotic grasping has been studied increasingly for a few decades. While progress has been made in this field, robotic hands are still nowhere near the capability of human hands. However, in the past few years, the increase in computational power and the availability of commercial tactile sensors have made it easier to develop techniques that exploit the feedback from the hand itself, the sense of touch. The focus of this thesis lies in the use of this sense. The work described in this thesis focuses on robotic grasping from two different viewpoints: robotic systems and data-driven grasping. The robotic systems viewpoint describes a complete architecture for the act of grasping and, to a lesser extent, more general manipulation. Two central claims that the architecture was designed for are hardware independence and the use of sensors during grasping. These properties enables the use of multiple different robotic platforms within the architecture. Secondly, new data-driven methods are proposed that can be incorporated into the grasping process. The first of these methods is a novel way of learning grasp stability from the tactile and haptic feedback of the hand instead of analytically solving the stability from a set of known contacts between the hand and the object. By learning from the data directly, there is no need to know the properties of the hand, such as kinematics, enabling the method to be utilized with complex hands. The second novel method, probabilistic grasping, combines the fields of tactile exploration and grasp planning. By employing well-known statistical methods and pre-existing knowledge of an object, object properties, such as pose, can be inferred with related uncertainty. This uncertainty is utilized by a grasp planning process which plans for stable grasps under the inferred uncertainty.

Kinematic gait analyses in healthy Golden Retrievers

Kinematic analysis relates to the relative movement between rigid bodies and finds application in gait analysis and other body movements, interpretation of their data when there is change, determines the choice of treatment to be instituted. The objective of this study was to standardize the march of Dog Golden Retriever Healthy to assist in the diagnosis and treatment of musculoskeletal disorders. We used a kinematic analysis system to analyse the gait of seven dogs Golden Retriever, female, aged between 2 and 4 years, weighing 21.5 to 28 kg, clinically normal. Flexion and extension were described for shoulder, elbow, carpal, hip, femorotibialis and tarsal joints. The gait was characterized lateral and had accepted hypothesis of normality for all variables, except for the stance of hip and elbow, considering a confidence level of 95%, significance level α = 0.05. Variations have been attributed to displacement of the stripes during movement and the duplicated number of reviews. The kinematic analysis proved to be a consistent method of evaluation of the movement during canine gait and the data can be used in the diagnosis and evaluation of canine gait in comparison to other studies and treatment of dogs with musculoskeletal disorders.

Kinematic analysis and modelling of biped locomotion systems

This paper presents the kinematic study of robotic biped locomotion systems. The main purpose is to determine the kinematic characteristics and the system performance during walking. For that objective, the prescribed motion of the biped is completely characterised in terms of five locomotion variables: step length, hip height, maximum hip ripple, maximum foot clearance and link lengths. In this work, we propose four methods to quantitatively measure the performance of the walking robot: energy analysis, perturbation analysis, lowpass frequency response and locomobility measure. These performance measures are discussed and compared in determining the robustness and effectiveness of the resulting locomotion.

A Design Methodology for the Compensation of Positioning Deviation in Gantry Manipulators

This work presents recent results concerning a design methodology used to estimate the positioning deviation for a gantry (Cartesian) manipulator, related mainly to structural elastic deformation of components during operational conditions. The case-study manipulator is classified as gantry type and its basic dimensions are 1,53m x 0,97m x 1,38m. The dimensions used for the calculation of effective workspace due to end-effector path displacement are: 1m x 0,5m x 0,5m. The manipulator is composed by four basic modules defined as module X, module Y, module Z and terminal arm, where is connected the end-effector. Each module controlled axis performs a linear-parabolic positioning movement. The planning path algorithm has the maximum velocity and the total distance as input parameters for a given task. The acceleration and deceleration times are the same. Denavit-Hartemberg parameterization method is used in the manipulator kinematics model. The gantry manipulator can be modeled as four rigid bodies with three degrees-of-freedom in translational movements, connected as an open kinematics chain. Dynamic analysis were performed considering inertial parameters specification such as component mass, inertia and center of gravity position of each module. These parameters are essential for a correct manipulator dynamic modelling, due to multiple possibilities of motion and manipulation of objects with different masses. The dynamic analysis consists of a mathematical modelling of the static and dynamic interactions among the modules. The computation of the structural deformations uses the finite element method (FEM).

Proposal of methodology for the modeling and control of manipulators

Industrial applications demand that robots operate in agreement with the position and orientation of their end effector. It is necessary to solve the kinematics inverse problem. This allows the displacement of the joints of the manipulator to be determined, to accomplish a given objective. Complete studies of dynamical control of joint robotics are also necessary. Initially, this article focuses on the implementation of numerical algorithms for the solution of the kinematics inverse problem and the modeling and simulation of dynamic systems. This is done using real time implementation. The modeling and simulation of dynamic systems are performed emphasizing off-line programming. In sequence, a complete study of the control strategies is carried out through the study of several elements of a robotic joint, such as: DC motor, inertia, and gearbox. Finally a trajectory generator, used as input for a generic group of joints, is developed and a proposal of the controller's implementation of joints, using EPLD development system, is presented.

Optimal design of 3R manipulators by using classical techniques and simulated annealin

In this paper, the optimum design of 3R manipulators is formulated and solved by using an algebraic formulation of workspace boundary. A manipulator design can be approached as a problem of optimization, in which the objective functions are the size of the manipulator and workspace volume; and the constrains can be given as a prescribed workspace volume. The numerical solution of the optimization problem is investigated by using two different numerical techniques, namely, sequential quadratic programming and simulated annealing. Numerical examples illustrate a design procedure and show the efficiency of the proposed algorithms.

Dynamical and Chemical Structure of Galactic Halos

This doctoral dissertation presents studies of the formation and evolution of galaxies, through observations and simulations of galactic halos. The halo is the component of galaxies which hosts some of the oldest objects we know of in the cosmos; it is where clues to the history of galaxies are found, for example, by how the chemical structure is related to the dynamics of objects in the halo. The dynamical and chemical structure of halos, both in the Milky Way’s own halo, and in two elliptical galaxies, is the underlying theme in the research. I focus on the density falloff and chemistry of the two external halos, and on the dynamics, density falloff, and chemistry of the Milky Way halo. I first study galactic halos via computer simulations, to test the long- term stability of an anomalous feature recently found in kinematics of the Milky Way’s metal-poor stellar halo. I find that the feature is transient, making its origin unclear. I use a second set of simulations to test if an initially strong relation between the dynamics and chemistry of halo glob-ular clusters in a Milky Way-type galaxy is affected by a merging satellite galaxy, and find that the relation remains strong despite a merger in which the satellite is a third of the mass of the host galaxy. From simulations, I move to observing halos in nearby galaxies, a challenging procedure as most of the light from galaxies comes from the disk and bulge components as opposed to the halo. I use Hubble Space Tele scope observations of the halo of the galaxy M87 and, comparing to similar observations of NGC 5128, find that the chemical structure of the inner halo is similar for both of these giant elliptical galaxies. I use Very Large Telescope observations of the outer halo of NGC 5128 (Centaurus A) and, because of the difficultly in resolving dim extragalac- tic stellar halo populations, I introduce a new technique to subtract the contaminating background galaxies. A transition from a metal-rich stellar halo to a metal-poor has previously been discovered in two different types of galaxies, the disk galaxy M31 and the classic elliptical NGC 3379. Unexpectedly, I discover in this third type of galaxy, the merger remnant NGC 5128, that the density of metal-rich and metal-poor halo stars falls at the same rate within the galactocentric radii of 8 − 65 kpc, the limit of our observations. This thesis presents new results which open opportunities for future investigations.

Stiffness based trajectory planning and feedforward based vibration suppression control of parallel robot machines

The dissertation proposes two control strategies, which include the trajectory planning and vibration suppression, for a kinematic redundant serial-parallel robot machine, with the aim of attaining the satisfactory machining performance. For a given prescribed trajectory of the robot's end-effector in the Cartesian space, a set of trajectories in the robot's joint space are generated based on the best stiffness performance of the robot along the prescribed trajectory. To construct the required system-wide analytical stiffness model for the serial-parallel robot machine, a variant of the virtual joint method (VJM) is proposed in the dissertation. The modified method is an evolution of Gosselin's lumped model that can account for the deformations of a flexible link in more directions. The effectiveness of this VJM variant is validated by comparing the computed stiffness results of a flexible link with the those of a matrix structural analysis (MSA) method. The comparison shows that the numerical results from both methods on an individual flexible beam are almost identical, which, in some sense, provides mutual validation. The most prominent advantage of the presented VJM variant compared with the MSA method is that it can be applied in a flexible structure system with complicated kinematics formed in terms of flexible serial links and joints. Moreover, by combining the VJM variant and the virtual work principle, a systemwide analytical stiffness model can be easily obtained for mechanisms with both serial kinematics and parallel kinematics. In the dissertation, a system-wide stiffness model of a kinematic redundant serial-parallel robot machine is constructed based on integration of the VJM variant and the virtual work principle. Numerical results of its stiffness performance are reported. For a kinematic redundant robot, to generate a set of feasible joints' trajectories for a prescribed trajectory of its end-effector, its system-wide stiffness performance is taken as the constraint in the joints trajectory planning in the dissertation. For a prescribed location of the end-effector, the robot permits an infinite number of inverse solutions, which consequently yields infinite kinds of stiffness performance. Therefore, a differential evolution (DE) algorithm in which the positions of redundant joints in the kinematics are taken as input variables was employed to search for the best stiffness performance of the robot. Numerical results of the generated joint trajectories are given for a kinematic redundant serial-parallel robot machine, IWR (Intersector Welding/Cutting Robot), when a particular trajectory of its end-effector has been prescribed. The numerical results show that the joint trajectories generated based on the stiffness optimization are feasible for realization in the control system since they are acceptably smooth. The results imply that the stiffness performance of the robot machine deviates smoothly with respect to the kinematic configuration in the adjacent domain of its best stiffness performance. To suppress the vibration of the robot machine due to varying cutting force during the machining process, this dissertation proposed a feedforward control strategy, which is constructed based on the derived inverse dynamics model of target system. The effectiveness of applying such a feedforward control in the vibration suppression has been validated in a parallel manipulator in the software environment. The experimental study of such a feedforward control has also been included in the dissertation. The difficulties of modelling the actual system due to the unknown components in its dynamics is noticed. As a solution, a back propagation (BP) neural network is proposed for identification of the unknown components of the dynamics model of the target system. To train such a BP neural network, a modified Levenberg-Marquardt algorithm that can utilize an experimental input-output data set of the entire dynamic system is introduced in the dissertation. Validation of the BP neural network and the modified Levenberg- Marquardt algorithm is done, respectively, by a sinusoidal output approximation, a second order system parameters estimation, and a friction model estimation of a parallel manipulator, which represent three different application aspects of this method.

Pharmacology of human experimental anxiety

This review covers the effect of drugs affecting anxiety using four psychological procedures for inducing experimental anxiety applied to healthy volunteers and patients with anxiety disorders. The first is aversive conditioning of the skin conductance responses to tones. The second is simulated public speaking, which consists of speaking in front of a video camera, with anxiety being measured with psychometric scales. The third is the Stroop Color-Word test, in which words naming colors are painted in the same or in a different shade, the incongruence generating a cognitive conflict. The last test is a human version of a thoroughly studied animal model of anxiety, fear-potentiated startle, in which the eye-blink reflex to a loud noise is recorded. The evidence reviewed led to the conclusion that the aversive conditioning and potentiated startle tests are based on classical conditioning of anticipatory anxiety. Their sensitivity to benzodiazepine anxiolytics suggests that these models generate an emotional state related to generalized anxiety disorder. On the other hand, the increase in anxiety determined by simulated public speaking is resistant to benzodiazepines and sensitive to drugs affecting serotonergic neurotransmission. This pharmacological profile, together with epidemiological evidence indicating its widespread prevalence, suggests that the emotional state generated by public speaking represents a species-specific response that may be related to social phobia and panic disorder. Because of scant pharmacological data, the status of the Stroop Color-Word test remains uncertain. In spite of ethical and economic constraints, human experimental anxiety constitutes a valuable tool for the study of the pathophysiology of anxiety disorders.

Control of Robot by Means of Human Thought

Brain computer interface (BCI) is a kind of human machine interface, which provides a new interaction method between human and computer or other equipment. The most significant characteristic of BCI system is that its control input is brain electrical activities acquired from the brain instead of traditional input such as hands or eyes. BCI technique has rapidly developed during last two decades and it has mainly worked as an auxiliary technique to help the disable people improve their life qualities. With the appearance of low cost novel electrical devices such as EMOTIV, BCI technique has been applied to the general public through many useful applications including video gaming, virtual reality and virtual keyboard. The purpose of this research is to be familiar with EMOTIV EPOC system and make use of it to build an EEG based BCI system for controlling an industrial manipulator by means of human thought. To build a BCI system, an acquisition program based on EMOTIV EPOC system is designed and a MFC based dialog that works as an operation panel is presented. Furthermore, the inverse kinematics of RV-3SB industrial robot was solved. In the last part of this research, the designed BCI system with human thought input is examined and the results indicate that the system is running smoothly and displays clearly the motion type and the incremental displacement of the motion.

Synergistic control of forearm based on accelerometer data and artificial neural networks

In the present study, we modeled a reaching task as a two-link mechanism. The upper arm and forearm motion trajectories during vertical arm movements were estimated from the measured angular accelerations with dual-axis accelerometers. A data set of reaching synergies from able-bodied individuals was used to train a radial basis function artificial neural network with upper arm/forearm tangential angular accelerations. The trained radial basis function artificial neural network for the specific movements predicted forearm motion from new upper arm trajectories with high correlation (mean, 0.9149-0.941). For all other movements, prediction was low (range, 0.0316-0.8302). Results suggest that the proposed algorithm is successful in generalization over similar motions and subjects. Such networks may be used as a high-level controller that could predict forearm kinematics from voluntary movements of the upper arm. This methodology is suitable for restoring the upper limb functions of individuals with motor disabilities of the forearm, but not of the upper arm. The developed control paradigm is applicable to upper-limb orthotic systems employing functional electrical stimulation. The proposed approach is of great significance particularly for humans with spinal cord injuries in a free-living environment. The implication of a measurement system with dual-axis accelerometers, developed for this study, is further seen in the evaluation of movement during the course of rehabilitation. For this purpose, training-related changes in synergies apparent from movement kinematics during rehabilitation would characterize the extent and the course of recovery. As such, a simple system using this methodology is of particular importance for stroke patients. The results underlie the important issue of upper-limb coordination.