Insect-inspired high-speed motion vision system for robot control

The mechanism for motion detection in a fly’s vision system, known as the Reichardt correlator, suffers from a main shortcoming as a velocity estimator: low accuracy. To enable accurate velocity estimation, responses of the Reichardt correlator to image sequences are analyzed in this paper. An elaborated model with additional preprocessing modules is proposed. The relative error of velocity estimation is significantly reduced by establishing a real-time response-velocity lookup table based on the power spectrum analysis of the input signal. By exploiting the improved velocity estimation accuracy and the simple structure of the Reichardt correlator, a high-speed vision system of 1?kHz is designed and applied for robot yaw-angle control in real-time experiments. The experimental results demonstrate the potential and feasibility of applying insect-inspired motion detection to robot control.     

Massively parallel implementations of theories for apparent motion

Two solutions for the correspondence problem for long-range motion are investigated. The first is a modification of the Minimal Mapping Theory (S. Ullman: The Interpretation of Visual Motion, MIT Press, Cambridge, 1979) that is implemented by a massively parallel network. In this network, every two units are interconnected, and thus, its convergence is fast and relatively independent of the number of image features. Computer simulations show that our method accounts as well as the Minimal Mapping Theory for apparent-motion phenomena, although some differences exist. Mathematical proofs provide conditions for the convergence of the network. The second 'solution' for the correspondence problem is called the Structural Theory. This theory assumes that the three-dimensional structure of viewed objects does not change fast in time. Then, the theory looks for the correspondence and three-dimensional structure that best fulfill this assumption. A massively parallel network implementation of this theory is also possible. However, its performance is poor due to the high complexity of its solution space. This supports Ullman's (1979) suggestion that the visual system separates the structure-from-motion process into two stages. First, a stage for motion measurement, and then a stage for structure recovery.     

Modeling of a bipedal robot using mutually coupled Rayleigh oscillators

The objective of the work presented here was the modeling of a bipedal robot using a central pattern generator (CPG) formed by a set of mutually coupled Rayleigh oscillators. We analyzed a 2D model, with the three most important determinants of gait, that performs only motions parallel to the sagittal plane. Using oscillators with integer relation of frequency, we determined the transient motion and the stable limit cycles of the network formed by the three oscillators, showing the behavior of the knee angles and the hip angle. A comparison of the plotted graphs revealed that the system provided excellent results when compared to experimental analysis. Based on the results of the study, we come to the conclusion that the use of mutually coupled Rayleigh oscillators can represent an excellent method of signal generation, allowing their application for feedback control of a walking machine.Acknowledgements The authors would like to express their gratitude to CNPq and CAPES for the financial support provided during the course of this research.     

Full-state tracking control of a mobile robot using neural networks

In this paper a nonholonomic mobile robot with completely unknown dynamics is discussed. A mathematical model has been considered and an efficient neural network is developed, which ensures guaranteed tracking performance leading to stability of the system. The neural network assumes a single layer structure, by taking advantage of the robot regressor dynamics that expresses the highly nonlinear robot dynamics in a linear form in terms of the known and unknown robot dynamic parameters. No assumptions relating to the boundedness is placed on the unmodeled disturbances. It is capable of generating real-time smooth and continuous velocity control signals that drive the mobile robot to follow the desired trajectories. The proposed approach resolves speed jump problem existing in some previous tracking controllers. Further, this neural network does not require offline training procedures. Lyapunov theory has been used to prove system stability. The practicality and effectiveness of the proposed tracking controller are demonstrated by simulation and comparison results.     

Quantifying biomechanical motion using Procrustes motion analysis

The ability to quantify and compare the movements of organisms is a central focus of many studies in biology, anthropology, biomechanics, and ergonomics. However, while the importance of functional motion analysis has long been acknowledged, quantitative methods for identifying differences in motion have not been widely developed. In this article, we present an approach to the functional analysis of motion and quantification of motion types. Our approach, Procrustes Motion Analysis (PMA) can be used to distinguish differences in cyclical, repeated, or goal-directed motions. PMA exploits the fact that any motion can be represented by an ordered sequence of postures exhibited throughout the course of a motion. Changes in posture from time step to time step form a trajectory through a multivariate data space, representing a specific motion. By evaluating the size, shape, and orientation of these motion trajectories, it is possible to examine variation in motion type within and among groups or even with respect to continuous variables. This represents a significant analytical advance over current approaches. Using simulated and digitized data representing cyclical, repeated and goal-directed motions, we show that PMA correctly identifies distinct motion tasks in these data sets.     

An in vitro study of implant--tooth-supported connections using a robot test system.

Many unsolved problems in dental implant research concern the interfacial stress distributions between the implant components, as well as between the implant surface and contacting bone. To obtain a mechanical understanding of how vertical and horizontal occlusal forces are distributed in this context, it is crucial to develop in vitro testing systems to measure the force transmission between dental implants and attached prostheses. A new approach to such testing, involving a robotic system, is described in this investigation. The system has been designed to produce simulated mandibular movements and occlusal contact forces so that various implant designs and procedures can be thoroughly tested and evaluated before animal testing or human clinical trials. Two commonly used fixed prosthesis designs used to connect an implant and a tooth, a rigid connection and a nonrigid connection, were fabricated and used for experimental verification. The displacement and force distributions generated during simulated chewing activities were measured in vitro. Force levels, potentially harmful to human bone surrounding the connected dental implant and tooth, were analyzed. These results are useful in the design of prostheses and connecting components that will reduce failures and limit stress transfer to the implant/bone interface.     

Iterative cooperation between parallel pathways for object and background motion

In a typical visual scene, one or more objects move relative to a larger background, which can itself be in motion as a result of the observer’s eyes moving with respect to the outside world. Here we show that accurate estimation of the background motion from an image velocity field can be accomplished through an iterative cooperation between two modules: one that specializes in calculating a weighted average velocity and another one calculating a velocity contrast map. We build on our analysis to provide a model for the tectum-pretectum loop in the nonmammalian midbrain. Our model accounts for some of the known properties of the tectal neurons (sensitivity to relative motion) and pretectal neurons (sensitivity to whole-field motion). It also agrees with our knowledge of the pretectotectal projection (divergent and inhibitory), and with the results of lesion studies in which the pretectal input to the tectum was removed, leading to hyperactivity of the tectal neurons and the animal. Our model also makes a testable prediction regarding the tectopretectal projection, i.e., that the presence of a larger object and a bigger discrepancy between the directions of motion for the object and the background lead to a larger error by the pretectum in estimating the background motion when the tectal input is abolished.     

In vivo pons motion within the skull

Finite element (FE) models are used to identify head injury mechanisms and design new and improved injury prevention schemes. Although brain-skull boundary conditions strongly influence the model mechanical responses, limited experimental data are available to develop an informed representation. We hypothesize that the spinal cord tension and gravity contribute to the pons displacement in vivo. Static high-resolution T1-weighted sagittal MR images of the inferior portion of the head in neutral and flexion positions were acquired in 15 human volunteers in both supine and prone postures. Boundaries of the pons and clivus were extracted with a gradient-based algorithm, and the pontes were fitted into ellipses. Assuming rigid body motion of the skull, image pairs in different postures were co-registered with an autocorrelation technique. By comparing images before and after the motion, we found that while the rotation of the pons is negligible relative to the skull, the pons displaces significantly at the foramen magnum, on the order of approximately 2 mm. When the spinal cord tension and gravity act in concert, the pons moves caudally; when opposed, superiorly, such that the influence of gravity on the pons is six times that of the spinal cord tension. Based on these findings, we recommend that the brainstem-skull interface be treated as a sliding (with or without friction) boundary condition in FE models of the human head.     

A study on large radial motion of arteries in vivo

This study analyses the radial periodic motion of an artery which is modelled as a thin cylinder of uniform cross-section subjected to dynamic inner pressure using the theory of finite deformation of elastic materials. The arterial tissue properties (anisotropy, homogeneity and incompressibility) are taken into account in an analysis based on the use of the strain energy function. The validity of the mathematical analysis is illustrated through numerical computation applying the available in vivo data for elastic constants of the canine middle descending thoracic aorta to the expressions for the intramural pressure and circumferential stresses obtained by solving the necessary equation of motion together with the boundary conditions. Results obtained in this study indicate very low stresses which suggest that the arteriosclerosis resulting from high stress gradients is effectively ruled out in this model.     

Prediction of total knee motion using a three-dimensional computer-graphics model

Twenty-three knees were sectioned, digitized, and standardized to determine the 'average' three-dimensional bony geometry and ligamentous attachments. Data on normal knee motion were obtained from a cadaveric study. An algorithm was written to simulate three-dimensional patella motion. Verification of the knee model was achieved by determining femoro-tibial and patello-femoral contact locations, as well as ligament length patterns, and comparing the results with published data. The criterion for maximum predicted knee motion with a prosthesis in place was the length of the posterior cruciate ligament. Three total knee replacement surfaces were mathematically generated: flat, laxity and conforming. A greater flexion angle was obtained with a flat tibial surface than for the laxity or conforming. Posterior tibial component displacement increased the range of motion, but only slightly. For all tibial surfaces, increased range of motion was achieved with a 10 degrees posterior tilt of the tibial tray. Anterior femoral component displacement increased motion due to reduction in posterior cruciate tension during flexion. The results are applicable to the design and surgical technique of total knee replacement.     

Biodegradation process development using a bacterial cytochrome in vivo

Pseudomonas putida PpG786 that contains the inducible enzyme system cytochrome P-450(cam) is considered for use as specialized biomass fore detoxification of hazardous hydrocarbons. The test substrate 1,2-dibromochloropropane (DBCP) is used to assess the organohalide degradation activity of P. Putida PpG786. Activity was found to be a strong function of intracellular heme content, variables which affect the culturing and processing of the cells, and oxygen tension in the degradation incubation medium. The lifetime for maintaing active biomass in chemostat washout operation, after including substrate was removed and then restarted, was also studied. These results indicate that initial activity of the P. Putida biomass is high enough, and decays slowly enough, so that industrial wastewater treatment at the operating conditions of a sequencing batch reactor (SBR) could remove hazardous compounds. (c) 1994 John Wiley & Sons, Inc.     

Characterization of macroautophagic flux in vivo using a leupeptin-based assay

Macroautophagy is a highly conserved catabolic process that is crucial for organ homeostasis in mammals. However, methods to directly measure macroautophagic activity (or flux) in vivo are limited. In this study we developed a quantitative macroautophagic flux assay based on measuring LC3b protein turnover in vivo after administering the protease inhibitor leupeptin. Using this assay we then characterized basal macroautophagic flux in different mouse organs. We found that the rate of LC3b accumulation after leupeptin treatment was greatest in the liver and lowest in spleen. Interestingly we found that LC3a, an ATG8/LC3b homologue and the LC3b-interacting protein p62 were degraded with similar kinetics to LC3b. However, the LC3b-related proteins GABARAP and GATE-16 were not rapidly turned over in mouse liver, implying that different LC3b homologues may contribute to macroautophagy via distinct mechanisms. Nutrient starvation augmented macroautophagic flux as measured by our assay, while refeeding the animals after a period of starvation significantly suppressed flux. We also confirmed that beclin 1 heterozygous mice had reduced basal macroautophagic flux compared to wild-type littermates. These results illustrate the usefulness of our leupeptin-based assay for studying the dynamics of macroautophagy in mice.     

Analysis of ubiquitination in vivo using a transgenic mouse model

The primary pathway for the proteolytic destruction of cellular proteins is through ubiquitin-mediated targeting to the proteasome. This pathway is pivotal not only in the elimination of damaged or misfolded proteins but also in the temporal, developmental, or signal-mediated destruction of normal cellular substrates. The list of known substrates of the ubiquitin/proteasome pathway is long, but most substrates have been identified in yeast or, more recently, in cultured mammalian cells. It is likely that many mammalian substrates with developmental or disease relevance have yet to be identified because their ubiquitination occurs in tissue or organ systems that cannot be adequately modeled in vitro. We have developed a transgenic mouse model that will allow the isolation and identification of these substrates. The human UbC promoter was used to drive expression of a hexahistidine-tagged version of human ubiquitin in a variety of mouse tissues from early embryonic stages, as assessed by a green fluorescent protein marker. Cleavage of the fusion protein by endogenous enzymes produced epitope-tagged ubiquitin that was detected both in monomeric form and conjugated to cellular proteins. This mouse model should facilitate in the analysis of normal and disease-related ubiquitination events in vivo.     

Localization of isoketal adducts in vivo using a single-chain antibody

Isoketals are highly reactive gamma-ketoaldehydes formed by the oxidation of arachidonic acid that rapidly adduct to proteins. To investigate the formation of isoketal adducts in vivo, we isolated and characterized a single-chain antibody from a phage displayed recombinant ScFv library that bound a model peptide adducted with synthetic 15-E2-isoketal. Recognition of isoketal adduct by this anti-isoketal adduct single-chain antibody was essentially independent of the amino acid sequence of adducted peptides or proteins. The antibody did not cross-react with 4-hydroxynonenal or 4-oxononanal adducts or with 15-F2t-isoprostane (8-iso-prostaglandin F2alpha). We investigated the formation of isoketal adducts in a well-established model of oxidative injury, hyperoxia. Exposure to >98% oxygen for 7 h dramatically increased both the number of immunoreactive airway epithelial cells and the intensity of immunoreactivity compared with animals exposed to normal room air (21% oxygen). We conclude that isoketal adducts form in epithelial cells as a result of high oxygen exposure and that this single-chain antibody provides a valuable tool to localize the formation of isoketal adducts in tissues in vivo.     

Characterization of macroautophagic flux in vivo using a leupeptin-based assay

Macroautophagy is a highly conserved catabolic process that is crucial for organ homeostasis in mammals. However, methods to directly measure macroautophagic activity (or flux) in vivo are limited. In this study we developed a quantitative macroautophagic flux assay based on measuring LC3b protein turnover in vivo after administering the protease inhibitor leupeptin. Using this assay we then characterized basal macroautophagic flux in different mouse organs. We found that the rate of LC3b accumulation after leupeptin treatment was greatest in the liver and lowest in spleen. Interestingly we found that LC3a, an ATG8/LC3b homologue and the LC3b-interacting protein p62 were degraded with similar kinetics to LC3b. However, the LC3b-related proteins GABARAP and GATE-16 were not rapidly turned over in mouse liver, implying that different LC3b homologues may contribute to macroautophagy via distinct mechanisms. Nutrient starvation augmented macroautophagic flux as measured by our assay, while refeeding the animals after a period of starvation significantly suppressed flux. We also confirmed that beclin 1 heterozygous mice had reduced basal macroautophagic flux compared to wild-type littermates. These results illustrate the usefulness of our leupeptin-based assay for studying the dynamics of macroautophagy in mice.     

Espin cross-links cause the elongation of microvillus-type parallel actin bundles in vivo

The espin actin-bundling proteins, which are the target of the jerker deafness mutation, caused a dramatic, concentration-dependent lengthening of LLC-PK1-CL4 cell microvilli and their parallel actin bundles. Espin level was also positively correlated with stereocilium length in hair cells. Villin, but not fascin or fimbrin, also produced noticeable lengthening. The espin COOH-terminal peptide, which contains the actin-bundling module, was necessary and sufficient for lengthening. Lengthening was blocked by 100 nM cytochalasin D. Espin cross-links slowed actin depolymerization in vitro less than twofold. Elimination of an actin monomer-binding WASP homology 2 domain and a profilin-binding proline-rich domain from espin did not decrease lengthening, but made it possible to demonstrate that actin incorporation was restricted to the microvillar tip and that bundles continued to undergo actin treadmilling at approximately 1.5 s-1 during and after lengthening. Thus, through relatively subtle effects on actin polymerization/depolymerization reactions in a treadmilling parallel actin bundle, espin cross-links cause pronounced barbed-end elongation and, thereby, make a longer bundle without joining shorter modules.