Eui- Jung

Design and motion planning of a two-moduled indoor pipeline inspection robot

This paper deals with design and motion planning of a reconfigurable robot that can be used for inspection of 80-100 mm pipelines. This robot consists of two connecting modules and each module consists of three pairs of caterpillar, which is operated by micro DC motor. The robot is foldable by using an embedded four-bar mechanism and compression of a spring connected to the four-bar allows the robot to maintain contact with the wall of pipelines. Controlling the speed of each caterpillar independently provides a steering capability to go through elbow and T-branch. The robot system has been developed and the validity of this mechanism was proved by experimentation.

Design of a reconfigurable indoor pipeline inspection robot

Recently, many pipeline inspection robot systems have been developed. We develop a reconfigurable robot that can be used for 80~100 mm pipeline. First of all, we derive the kinematics of this mechanism and determine the actuator size by static analysis. The mechanism consists of three pairs of caterpillar, each of which is operated by a micro DC motor. The robot is designed foldable when each chain contacts the wall of pipelines. Thus, the robot can be operated in various sizes of pipeline. Controlling the speed of caterpillar independently provides a steering capability. So it can go through elbow and T-branch. The robot system has been developed and the validity of this mechanism was proved by experimentation.

JAIST Robotic Walker control based on a two-layered Kalman filter

This paper presents a new control scheme of JAIST Active Robotic Walker (JARoW) developed to provide elderly people with sufficient ambulatory capability. Toward its practical use, our focus is placed on how to allow easier and reliable maneuverability by creating a natural user interface. Specifically, our challenge lies in providing a well-functioning controller by detecting what the user wants to do or their intentions. A Kalman filter based tracking scheme is realized to estimate and predict the locations of the users legs and body in real time. The feedback control can then adjust the motions of JARoW corresponding to the actual users walking behaviors. Our experiments confirm that JARoW can autonomously adjust its motion direction and velocity without requiring any additional control inputs.

Localization algorithm based on Zigbee wireless sensor network with application to an active shopping cart

This paper describes a localization algorithm based on hybrid sensor system with application to an active shopping cart. For a given experimental environment, the probability localization method is applied to confirm the global coordinate of the target customer. And a hybrid sensor system combining the Zigbee and odometry is used to improve the localization performance of the active shopping cart. The shopping cart is equipped with motors for mobility and sensors for tracking person. Through experimental work, we corroborate the feasibility of the proposed localization algorithm.

SLAM in indoor pipelines with 15mm diameter

As the need of well-being life becomes popular, one of important issues is clean drinking water. Delivery of clean water from the main water fountain to the end user is very crucial. However, typical pipelines are being contaminated by rust and several chemicals. And thus the water being delivered to house is not trustful as the drinking water. Thus, cleaning and inspection of the in-house pipeline becomes an important issue. However, it is hard to find a small-sized motor adequate to the robot for the inspection of the in-house pipeline with diameter of 15 mm. Thus, this paper introduces a new semi-automatic pipeline inspection robotic system for small-sized pipelines. Three different types of inspection robot system were developed, which have different combination of sensors to realize SLAM. The robot system is carried by an extension cable, which is in advance being penetrated into the pipeline by a compressed air. The feasibility of the proposed SLAM using this semi-automatic pipeline inspection robotic system was verified through experimentation.

Mobile shopping cart application using kinect

This paper presents a novel application of a perception sensor Kinect to a mobile shopping cart (MSC). We attempt to exercise daily-used instructions as commands by using skeleton tracking of the Kinect sensor. A pair of differential-driven wheels together with dynamical system is mounted on the chassis of the shopping cart, which reshapes the shopping cart as a mobile robot. The commercialized Kinect sensor is employed to detect the gestures of the human body. The mobile shopping cart equipped with Kinect could execute different gesture command, for example, go ahead, draw back, set stop and so on. The response time is within 0.5 second. To do the right reflexes, we define a set of trajectory determined by the gesture of the target customer. The experiment result shows the validity of proposed method.

Motion planning algorithms of an omni-directional mobile robot with active caster wheels

This work deals with motion planning algorithms of an omni-directional mobile robot with active caster wheels. A typical problem occurred in the motion control of such omni-directional mobile robot, which has been identified through experimental experiences, is skidding of the mobile wheel. It sometimes results in uncertain rotation of the steering wheel. To cope with this problem, a motion planning algorithm which resolves the skidding problem and uncertain motions of the steering wheel is mainly investigated. For navigation of the mobile robot, the posture of the omni-directional mobile robot is initially calculated using the odometry information. Then, the accuracy of the mobile robot’s odometry is measured through comparison of the odometry information and the external sensor measurement. Finally, for successful navigation of the mobile robot, a motion planning algorithm that employs kinematic redundancy resolution method is proposed. Through simulations and experimentation, the feasibility of proposed algorithms was verified.

Task-oriented navigation algorithms for an outdoor environment with colored borders and obstacles

This paper presents task-oriented navigation algorithms used for an outdoor environment. The goals of the navigation are recognizing colored border lines on both sides of a path, avoiding obstacles on the path, and navigating the given path. To recognize the colored border lines with one camera, we apply a support vector data description method, which employs six color features extracted from two color models. To avoid collision with obstacles on the path, we fuse the data of the lines measured by a camera and the obstacles measured by a laser range finder. These algorithms were applied to autonomous navigation of about 100 m long curved track. We demonstrate that a four-wheel skid-steering mobile robot successfully finishes the mission.

Navigation of an omni-directional mobile robot with active caster wheels

This work deals with navigation of an omni-directional mobile robot with active caster wheels. Initially, the posture of the omni-directional mobile robot is calculated by using the odometry information. Next, the position accuracy of the mobile robot is measured through comparison of the odometry information and the external sensor measurement. Finally, for successful navigation of the mobile robot, a motion planning algorithm that employs kinematic redundancy resolution method is proposed. Through experiments for multiple obstacles and multiple moving obstacles, the feasibility of the proposed navigation algorithm was verified.

Implementation of an embedded omni-directional mobile robot with active caster wheels

This work deals with a motion planning algorithm of an omni-directional mobile robot with active caster wheels. A typical problem occurred in the motion control, which has been identified through experimental experiences, is skidding of the mobile wheel. It sometimes results in uncertain rotation of the steering wheel. This is due to mismatching between the command and the mechanical hardware. To cope with this problem, a motion planning algorithm pursuing the natural configuration of the mobile wheel was mainly investigated, which resolves the skidding problem and uncertain motions of the steering wheel. Through simulations and experimentation, the feasibility of this algorithm was verified.