Image-based fuzzy trajectory tracking control for four-wheel steered mobile robots

A four-wheel steered mobile robot is fit for a higher power or improvement in the movement speed of a robot than a two-independent wheeled one. Since a steered mobile robot that slips very often cannot apply a popular dead-reckoning method using rotary encoders, it is desirable to use external sensors such as cameras. This paper describes a method to trace a straight line for four-wheel steered mobile robots using an image-based control method. Its controller is designed as a fuzzy controller and evaluated through some simulations and real robot.     

Vitamin B6 suppresses serine protease inhibitor 3 expression in the colon of rats and in TNF-α-stimulated HT-29 cells

Scope: Previous reports in the areas of animal studies and, recently epidemiology, have linked anti‐tumorigenic and anti‐inflammatory effects to dietary vitamin B6. This study investigated the molecular mechanism of these effects of vitamin B6. Methods and results: DNA microarray analysis was used to obtain information on changes in colon gene expression from vitamin B6 (pyridoxine) repletion in vitamin B6‐deficient rats. Pyridoxine supplementation down‐regulated the inflammatory molecule, serine protease inhibitor clade A member 3 (SPI‐3) mRNA expression in the colon. This study also showed that tumor necrosis factor α (TNF‐α) induced SPI‐3 mRNA expression in HT‐29 human colon cancer cells, and vitamin B6 (pyridoxal hydrochloride) pretreatment of HT‐29 cells inhibited TNF ‐induced mRNA expression of SPI‐3. Vitamin B6 inhibited TNF‐α‐induced NF‐κB activation via suppression of IκBα degradation in HT‐29 cells. HT‐29 cells stably expressing epitope‐tagged ubiquitin were generated and vitamin B6 pretreatment was shown to inhibit ubiquitination of the IkB protein in response to TNF‐α‐i. Conclusion: Vitamin B6 suppressed SPI‐3 expression in the colon of rats and in TNF‐α‐stimulated HT‐29 cells. Further, this study showed a possible role of vitamin B6 in the regulation of protein ubiquitination.     

A desktop NC machine tool with a position/force controller using a fine-velocity pulse converter

In this paper, a new desktop NC machine tool with the ability of compliance control is presented for finishing of metallic molds with small curved surface. The NC machine tool consists of a three-axis robot whose single one has a high position accuracy of 1 μm, which means that it can perform higher resolutions of position and force, compared to general industrial robots. A ball-end abrasive tool is attached to the tip of z-axis through a force sensor. In order to first confirm the application limit of an industrial robot to a finishing task, we evaluate the backlash that causes inaccuracy in position at the tip of the abrasive tool, by simply measuring the position and force. The basic resolution of the proposed NC machine tool is shown through the similar position and force measurement to one used in the conventional robot. Next, a CAD/CAM-based position/force control method with a fine-velocity pulse converter is proposed for a pulse-based servo controller that actuates the single-axis of the robot. We further introduce a systematic method for tuning a desired damping, which is coming from the critically damped condition used in a static force control system. The desired damping is the most important factor when an impedance model following force control is applied. Finally, to evaluate the characteristics of the NC machine tool, a profiling control experiment is conducted along a spiral path for a plastic lens mold with axis-symmetric concave areas. Consequently, it is confirmed that the proposed NC machine tool has a good performance, which can be applied to the finishing process of plastic lens molds.     

Feed-rate control using fuzzy reasoning for NC machine tools

In this article, a 3D design and machining system based on an NC machine tool with a rotary unit is first introduced to effectively produce attractive paint rollers. A post-processor is proposed to transform a base tool path called cutter location data (CL data) to NC data, mapping the y-directional pick feed to the rotational angle of the rotary unit. The 3D machining system with the postprocessor allows us easily to transcribe a relief design on a flat model surface to a cylindrical model surface. The postprocessor has another function that systematically adjusts the feed-rate according to the curvature of each design to prevent the machined surface from being chipped. The post-processor generates suitable feed-rate codes by using a simple fuzzy reasoning method while checking the edges in relief designs. Experimental results show that wooden paint rollers with attractive relief designs can be successfully machined without undesirable edge chipping. Next, the proposed fuzzy feed-rate generator is further applied to a mold-polishing robot, so that the polishing time can be reduced by about 30% compared with the case without the generator. This work was presented in part at the 12th International Symposium on Artificial Life and Robotics, Oita, Japan, January 25–27, 2007     

Depth perception for moving images shown on a large LED display with an aperture grille

Abstract— A type of depth illusion created by the use of an aperture grille is reported. When viewing a moving target through multiple slits, a movement with depth is perceived, which was originally reported (see Ref. 11). The binocular delay is considered to cause a virtual disparity between both perspective images with apparent movement. By using an LED display with an aperture grille as a stereoscopic display, perceived distance caused by a binocular delay has been measured. The measured distance is compared with the perceived distance for stereoscopic still images shown on a stereoscopic LED panel. The comparison supported that the binocular delay is converted into binocular disparity. Furthermore, pair‐comparison tests were conducted to investigate depth impressions. It was found that use of an aperture grille improves depth impression for a movie that was taken with a laterally moving camera.     

A Real-Time Kinematics on the Translational Crawl Motion of a Quadruped Robot

It is known that the kinematics of a quadruped robot is complex due to its topology and the redundant actuation in the robot. However, it is fundamental to compute the inverse and direct kinematics for the sophisticated control of the robot in real-time. In this paper, the translational crawl gait of a quadruped robot is introduced and the approach to find the solution of the kinematics for such a crawl motion is proposed. Since the resulting kinematics is simplified, the formulation can be used for the real-time control of the robot. The results of simulation and experiment shows that the present method is feasible and efficient.     

Analysis and Control for an Omnidirectional Mobile Manipulator

This paper describes analysis and control for a holonomic omnidirectional mobile manipulator, in which the holonomic omnidirectional platform consists of three lateral orthogonal wheel assemblies and a mounted manipulator with three rotational joints is located at the center of gravity of the platform. We first introduce the kinematic model for the mobile manipulator and derive the dynamical model by using the Newton–Euler method, where a model which simultaneously takes account of features of both the manipulator and the mobile parts is given to analyze the effect of the movement of mounted manipulator on the platform. Then, the computed torque control and the resolved acceleration control methods are used to show that the holonomic omnidirectional mobile manipulator can be controlled so as to retain any end-effector position and orientation, irrespective of the direction of external applied force. The validity of the model and the effectiveness of the present mobile manipulator are proved by using several numerical simulations and 3D animations.     

Adaptive learning with large variability of teaching signals for neural networks and its application to motion control of an industrial robot

Recently, various control methods represented by proportional-integral-derivative (PID) control are used for robotic control. To cope with the requirements for high response and precision, advanced feedforward controllers such as gravity compensator, Coriolis/centrifugal force compensator and friction compensators have been built in the controller. Generally, it causes heavy computational load when calculating the compensating value within a short sampling period. In this paper, integrated recurrent neural networks are applied as a feedforward controller for PUMA560 manipulator. The feedforward controller works instead of gravity and Coriolis/centrifugal force compensators. In the learning process of the neural network by using back propagation algorithm, the learning coefficient and gain of sigmoid function are tuned intuitively and empirically according to teaching signals. The tuning is complicated because it is being conducted by trial and error. Especially, when the scale of teaching signal is large, the problem becomes crucial. To cope with the problem which concerns the learning performance, a simple and adaptive learning technique for large scale teaching signals is proposed. The learning techniques and control effectiveness are evaluated through simulations using the dynamic model of PUMA560 manipulator.