Toshiyuki Asakura

Manipulator visual servoing and tracking of fish using a genetic algorithm

This paper presents the real‐time visual servoing of a manipulator and its tracking strategy of a fish, by employing a genetic algorithm (GA) and the unprocessed gray‐scale image termed here as “raw‐image”. The raw‐image is employed to shorten the control period, since it has more tolerance of contrast variations occurring within an object, and between one input image and the next one. GA is employed in a method called 1‐step‐GA evolution. In this way, for every generational step of the GA process, the found results, which express the deviation of the target in the camera frame, are output for control purposes. These results are then used to determine the control inputs of the PD‐type controller. Our proposed GA‐based visual servoing has been implemented in a real system, and the results have shown its effectiveness by successfully tracking a moving target fish.

Study of machine fault diagnosis system using neural networks

Develops a machine fault diagnosis system using neural networks and spectral analysis. A neural network is applied to the fault diagnosis of the machine. The neural network has learning and memory capability. By the learning of normal and abnormal states of the object system, a method with neural networks is proposed which can diagnose a fault of the machine. The proposed fault diagnosis system is based on the spectrum of vibrations or sounds obtained from the operating machine. The difference between normal and abnormal data becomes clearer when comparing time series data. It is suitable for the detection of the fault to utilize changes of spectral data. Using this method, it is shown that it can detect unknown fault patterns. Fault diagnosis experiments are performed on both a wood slicing machine and an electromagnetic valve. The possibility of an online fault diagnosis system is examined through the construction of an online data processing system for an electromagnetic valve and it is shown that the fault diagnosis can be performed in real time. Through these results, the effectiveness of the proposed fault diagnosis system is verified.

Robust scene recognition using a GA and real-world raw-image

Abstract This paper presents a new concept of scene recognition by a genetic algorithm (GA), using the 2-D gray-scale image of a working space, termed here as raw-image, and a model shaping the 2-D top-surface of a target object. In fact here, the problem of object recognition in the raw-image is changed into an optimization problem of a model-based evaluation function. We make use in this research of a GA, as a search and optimization method. This GA employs a model-based fitness function as its objective function to perform the search of a target in the raw-image. In this research, three object models, namely a frame model, a surface model, and a surface-strips model are investigated in order to determine which one is the best for scene recognition in a noisy environment. Also, in order to appraise the recognition performance of each model, a comparative study is performed by analyzing the answers to the following criteria questions: sensitivity, reliability, and speed. The effectiveness of the method has been verified through experiments using real-world raw-images, and the method has shown its robustness of object recognition with the surface-strips model, in spite of the noises in the scene.

Visual servoing to fish and catching using global/local GA search

This paper presents a vision related technique for a manipulator real-time visual servoing. The method utilizes the global search feature of a genetic algorithm (GA) and a local search technique of the GA and also the unprocessed gray-scale image called here as raw-image, in order to perform recognition of a known target object being imaged. Also in GA process, the computation of the fitness function is based on the configuration of an object model designated as surface-strips model. The raw-image is used since it is more tolerant of contrast variations from an input image to the next one, and moreover does not require any filtering processing time. The global GA is utilized together with the local GA in order to recognize the target shape and detect the position and orientation simultaneously, and to increase the GAs convergence speed so as to provide faster and better recognition results. In order to evaluate the effectiveness of the proposed scene recognition method, experiments to track a fish by hand-eye camera and catch the fish with a net attached at hand of the manipulator have been done. The success to catch has shown the effectiveness of the proposed technique for manipulator real-time visual servoing.

GA-pattern matching-based manipulator control system for real-time visual servoing

—In robotic applications, tasks of picking and placing are the most fundamental ones. Also, for a robot manipulator, the recognition of its working environment is one of the most important issues to do intelligent tasks, since this aptitude enables it to work in a variable environment. This paper presents a new control strategy for robot manipulators, which utilizes visual information to direct the manipulator in its working space, to pick up an object of known shape, but with arbitrary position and orientation. During the search for an object to be picked up, vision-based control by closed-loop feedback, referred to as visual servoing, is performed to obtain the motion control of the manipulator hand. The system employs a genetic algorithm (GA) and a pattern matching technique to explore the search space and exploit the best solutions by this search technique. The control strategy utilizes the found results of GA-pattern matching in every step of GA evolution to direct the manipulator towards the target ...

Real-time recognition of road traffic sign in moving scene image using new image filter

This research is concerned with a new technology for the recognition of traffic sign from a motion picture. In order to realize a real-time position recognition, the genetic algorithm (denoted by GA) with search region limits is proposed. Through experiments, it is shown that new technology is sufficiently valid on real-time recognition to objects of motion picture.

Evolutionary Scene Recognition and Simultaneous Position/Orientation Detection

This paper presents a new method of scene recognition for manipulator real-time visual servoing, which utilizes a hybrid genetic algorithm in combination with a model shaping a target of known shape, and the unprocessed gray-scale image of a scene, termed here as raw-image. The scene recognition method presented here is concerned with the simultaneous recognition of the shape and detection of the position and orientation in the two-dimensional raw-image, of a three-dimensional target being imaged. This approach of scene recognition is applied for the recognition of either a non-self moving target as well as for the a self-moving target such as a living animal, in the presence of noises like lighting condition variations and other objects in the scene, considered as noises. The raw-image is employed since it does not increase the original noise, thereby being more transportable, and moreover contrary to a binary image processing, does not require any filtering processing time. In fact here, the problem of an object recognition in the raw-image is changed to an optimization problem of a model-based evaluation function, named surface-strips model-based fitness function. This fitness function possesses information about the shape of a target, and consists in the computation of the brightness difference between an internal surface and a contour-strips. In this research, in order to find a target object in every newly input raw-image to the recognition system, the highest peak of the distribution of the surface-strips model-based fitness function, which corresponds to the recognition results of the designated target, is searched by a hybrid genetic algorithm, which employs a population of potential solutions to perform the search of the target. This hybrid genetic algorithm employs the “global” search feature of a two-point crossover genetic algorithm (GA), to search a target, together with a GA-based localized search technique that focuses on the target of interest found so far, in order to perform an intensive localized search. The localized GA search technique relies on mutation of bits on the lower portion of genes in positional and orientational binary strings, in order to improve the GA-based scene recognition performance, in terms of fast and reliable recognition of the target. This generational scene recognition by the hybrid genetic algorithm can be designated as “evolutionary scene recognition and position/orientation detection”. In order to appraise the proposed scene recognition method, experiments by a hand-eye camera of a robot manipulator have been conducted to show its robustness and reliability with respect to various disturbing objects and lighting condition changes, and its effectiveness to recognize a natural fish swimming in a pool. These results have shown the suitableness of the method for manipulator real-time visual servoing.

Machine intelligence of a mobile manipulator to utilize dynamically interfered motion and nonlinear friction

Dynamical interferences have been thought that they should be erased to improve control accuracy. However it may be possible to improve the performance of total motion using the interferences. We propose a method to acquire a kind of machine intelligence to utilize dynamically interfered motion. The machine intelligence is defined here as an ability that the machine can find by itself a way to use dynamical interferences and nonlinear friction to obtain a desired motion. We propose a strategy of how a machine uses the effects of the dynamical interferences, and how it acquires the way to achieve an objective motion. The desired motion is traveling of a 1-link mobile manipulator by using interfering motion of the mounted link, which does not possess driving motors nor brakes. The proposed method is composed of functions to give the machine sample motions using Fourier series and to improve the Fourier coefficients by evaluating the motion results based on a function used in a genetic algorithm as a fitness function. Further, an ability to avoid collisions between the mounted manipulator and the floor is added to the traveling ability to confirm that the proposed method could be adapted to many objectives. We confirmed by simulations and real experiments that the mobile manipulator could find effective motion that makes it travel forward without colliding against the floor.

Development of a new type of machining robot-a new type of driving mechanism

On the bases of an analysis of the reasons for the low kinematic accuracy and poor stiffness of the conventional robots, a new type of machining robot that is particularly adapted to perform cutting tasks is developed in this study. A new type of 4-axis combined driving arm is presented and its principle is firstly introduced in this paper, and the mechanical performance is analyzed. Analytical results show that the mechanism is characterized by high stiffness and good kinematic accuracy.

Position control and explicit force control of constrained motions of a manipulator for accurate grinding tasks

On the basis of an analysis of the interaction between a manipulator for a grinding process and a working object in the task space, motions of the constrained dynamic system of the robot are modeled first in this paper. In the model, the constrained forces are included and expressed as a function of the state and input generalized forces by using the equation of constraints. Using this result, a control law is proposed by taking advantage of the redundancy of input generalized forces to the constrained forces. A controller for the grinding robot is then constructed according to this control law and without involving any force sensors. Simulations have been done for justifying the feasibility of the controller by taking an articulated planar two-link manipulator as an example. Results show that the constrained force is explicitly controlled with the proposed control law and the effectiveness of the controller has been verified.