Young-Jae Ryoo

A neural network model of electric differential system for electric vehicle

This paper describes a neural network model for an electrical differential system for electric vehicle. When a vehicle drives along curved road lane, the speed of the inner wheel has to be different from that of the outer wheel in order to prevent the vehicle vibrating and traveling an unsteady path. Because each wheel of this electrical vehicle has an independent driving force, an electrical differential system is required to replace a gear differential system. However, it is difficult to analysis the nonlinear behavior of the differential system in relation to the vehicle speed and steering angle, as well as vehicle structure. Therefore, a neural network is used to learn the relationships. To realize the neural network model, the speed data was acquired for the inner wheel and outer wheel, using an experimental electric vehicle at various speeds and steering angles. With this information, the differential system can be controlled using a neural network model of the nonlinear relationships.

Development of steering control system for autonomous vehicle using array magnetic sensors

In this paper, development of steering control system for autonomous vehicle using array magnetic sensors of the magnetic guidance based is described. The Steering system is operated following on magnetic marker position of on the roadway. For experimental test, steering system is designed and operation test is realized.

Position Sensing System for Magnet Based Autonomous Vehicle and Robot Using 1-Dimensional Magnetic Field Sensor Array

In this paper, a position sensing system for magnet based autonomous vehicle and robot using 1-diemnsional magnetic field sensor array was described. In advanced vehicle and robot control, position sensing is an important task for the identification of their locations, such as the lateral position within a trajectory. The magnet based autonomous vehicle and robot was identified position using magnetic materials. In the magnetic sensing system, the Earth field is a disturbance. It should be estimated a real-time. And, the memory space should be reduced in implementation. To solve the above problems, this paper proposes the magnetic field sensor array system included a vertical component of magnetic field, a linear region of the sensor output, method of position determination using a simple equation with a microcontroller. The proposal is verified practicability for magnetic position sensing system by the experimental results

Development of an Omnidirectional Walking Engine for Full-Sized Lightweight Humanoid Robots

In this paper, we propose and demonstrate an omnidirectional walking engine that achieves stable walking using feedback from an inertial measurement unit (IMU). The 3D linear inverted pendulum model (3D-LIPM) is used as a simplified model of the robot, the zero moment point (ZMP) criterion is used as the stability criterion, and only the feedback from the IMU is utilized for stabilization. The proposed walking engine consists of two parts; the omnidirectional gait generator, and the stability controller. ZMP equations, derived based on the 3D-LIPM, are used in the omnidirectional gait generator. The solutions of the differential equations are directly used which reduces the computation cost compare to other existing methods. Two kinds of feedback controllers are implemented for the stability controller; one is the indirect reference ZMP controller, and the other is the indirect joint controller. The walking engine is tested on a lightweight, full-sized, 21-degree-of-freedom (DOF) humanoid robot CHARLI-L (Cognitive Humanoid Autonomous Robot with Learning Intelligence, version Lightweight) which stands 141 cm tall and weighs only 12.7 kg. The design goals of CHARLI-L are low development cost, lightweight, and simple design, which all match well with the proposed walking engine. The results of the experiments present the efficacy of our approach.Copyright

Design of magnet based position sensing system for autonomous vehicle robot

In this paper, a position sensing system using magnet and magnetic sensor for autonomous vehicle robot is designed. Magnetic sensing is a reliable technology that has been developed for the purposed of position measurement and guidance, especially for applications in autonomous vehicle robot. The magnetic fields from a sample magnet are first measured to determine the characteristics of the field patterns as the basis for detection and position identification. A discussion of a position sensing system with a rejection algorithm of background field is presented. The position sensing technique was implemented in the guidance of autonomous vehicle robot. The results of lateral motion control show that position-sensing system can be useful for an autonomous vehicle robot.

Classification of materials using temperature response curve fitting and fuzzy neural network

Abstract This paper describes a system that can be used to classify an unknown material regardless of the change of ambient temperature using temperature response curve fitting and fuzzy neural network (FNN). There are some problems to realize the classification system using temperature response. It requires too many memories to store the vast temperature response data and it has to be filtered to remove noise which occurs in experiment. And the temperature response is influenced by the change of ambient temperature. So, this paper proposes a practical method using curve fitting to remove above problems of memories and noise. And FNN is proposed to overcome the problem caused by the change of ambient temperature. Using the FNN which is learned by temperature responses on fixed ambient temperature and known thermal conductivity (TC), the TC of the material can be inferred on various ambient temperature. So the material can be classified by the TC.

Development of Torque Sensor with High Sensitivity for Joint of Robot Manipulator Using 4-Bar Linkage Shape.

The torque sensor is used to measure the joint torque of a robot manipulator. Previous research showed that the sensitivity and the stiffness of torque sensors have trade-off characteristics. Stiffness has to be sacrificed to increase the sensitivity of the sensor. In this research, a new torque sensor with high sensitivity (TSHS) is proposed in order to resolve this problem. The key idea of the TSHS comes from its 4-bar linkage shape in which the angular displacement of a short link is larger than that of a long link. The sensitivity of the torque sensor with a 4-bar link shape is improved without decreasing stiffness. Optimization techniques are applied to maximize the sensitivity of the sensor. An actual TSHS is constructed to verify the validity of the proposed mechanism. Experimental results show that the sensitivity of TSHS can be increased 3.5 times without sacrificing stiffness.

Motion Editing Tool to Create Dancing Motions of Humanoid Robot

In this paper, a motion editing tool to create dancing motions of a humanoid robot is proposed. In order to build performances or dancing of a humanoid robot, a motion editing tool to create specific motions is necessary. Especially, to generate more natural motions is required for a dancing robot. We proposed a motion editing tool and algorithm to create the natural motions. The proposed motion editing tool can create robots motions composed of several steps which are captured from every joint while the robot plays. The motion editing tool generates the continuous motion interpolated between each steps. A humanoid robot of 50 cm tall is developed to test the proposed tool. The robot using the motion editing tool was demonstrated the natural dancing performance.

Neural Network Mapping of Magnet Based Position Sensing System for Autonomous Robotic Vehicle

In this paper a neural network mapping of magnet based position sensing system for an autonomous robotic vehicle. The position sensing system using magnetic markers embedded under the surface of roadway pavement. An important role of magnetic position sensing is identification of vehicles location. The magnetic sensor measures lateral distance when the vehicle passes over the magnetic marker. California PATH has developed a table-look-up as an inverse map. But its requires too many memories to store the vast magnetic field data. Thus we propose the magnetic guidance system with simple mapping using neural network.

Neural Network Control for Visual Guidance System of Mobile Robot

This paper describes a neural network control for a visual guidance system of a mobile robot to follow a guideline. Without complicated geometric reasoning from the image of a guideline to the robot-centered representation of a birds eye view in conventional studies, the proposed system transfers the input of image information into the output of a steering angle directly. The neural network controller replaces the nonlinear relation of image information to a steering angle of robot on the real ground. For image information, the feature points of guideline are extracted from a camera image. In a straight and curved guideline, the driving performances by the proposed technology are measured in simulation and experimental test.