Search

Purpose Many manufacturing sites require precision assembly. Particularly, similar to cell phones, assembly at the sub-mm scale is not easy, even for humans. In addition, the system should assemble each part with adequate force and avoid breaking the circuits with excessive force. The purpose of this study is to assemble high precision components with relatively reasonable vision devices compared to previous studies. Design/methodology/approach This paper presents a vision-force guided precise assembly system using a force sensor and two charge coupled device (CCD) cameras without an expensive 3-dimensional (3D) sensor or computer-aided design model. The system accurately estimates 6 degrees-of-freedom (DOF) poses from a 2D image in real time and assembles parts with the proper force. Findings In this experiment, three connectors are assembled on a printed circuit board. This system obtains high accuracy under 1 mm and 1 degree error, which shows that this system is effective. Originality/value This is a new method for sub-mm assembly using only two CCD cameras and one force sensor.

This paper presents an appearance based place recognition method using omni-directional images which describe 360° panoramic scene around a camera at one time. We utilize a spectral graph matching method to measure a degree of similarity between omni-directional query and topological node images. This method builds an affinity matrix of a graph whose nodes represent the potential correspondences while the weights on the edges measure the similarity between pairs of potential correspondences. In this paper, we compute the elements of affinity matrix with individual photometric feature matching and geometrical graph shape matching between omni-directional images. The proposed method enables robust place recognition since it considers both the geometrical and photometrical similarities. The experimental results show that the proposed method can robustly estimate the place in the dynamic indoor environment.

This paper addresses a vision-based 3D motion estimation framework for humanoid robots, which copes with human-like walking pattern. A humanoid robot, called Roboray, is designed for dynamic walking control with heel-toe motion like a human. In spite of stability and energy efficiency of the dynamic walking, it accompanies larger swaying motion and more uncertainty in camera movement than the conventional ZMP (Zero Moment Point)-based walking does. The framework effectively uses on-board odometry information from the robot to improve the performance of the visionbased motion estimation. To accomplish this, we propose an onboard odometry filter which fuses kinematic odometry, visual odometry, and raw IMU data. And the odometry filter is combined with vision-based SLAM to provide accurate motion model, so it enhances the SLAM estimates. Experimental results in indoor environment verify that the framework can successfully estimate the pose of Roboray in real-time.

We propose a practical and efficient navigation method of the indoor mobile robot using 1D panoramic images. Mobile robot navigation, one of the most importation components in the robotic application, was carried out using the omni-directional camera that can capture 360deg images around a robot. Therefore, this camera has many advantages in the indoor navigation. In this paper, position of the robot can be estimated by 1D panoramic image correlations. This 1D image is the circular horizontal line in the omni-directional image. The proposed method can estimate the position of the robot without any previous position information in the short time i.e., kidnap problem can be coped with. The path of a robot is generated based on the node map that includes 1D panoramic images and the node information at the captured points. For the feasibility test of the proposed algorithms, we applied them to the mobile robot, iMARO-III. In this test, iMARO-III has succeeded in real world operation without any interaction with operator.

This paper proposes a traction control algorithm which can be independently implemented to each wheel without extra sensors and devices compared with standard speed control. The algorithm estimates stick-slip of wheels based on estimation of angular acceleration so that the traction force induced by torque of wheel converses between maximum static friction and kinetic friction. Simulations are performed to verify the validity of the algorithm. The proposed traction control algorithm obtained 40.5% reduction of total slip distance and 48.4% reduction of dissipated work on the contact point compared with standard speed control. Furthermore, the algorithm does not require complex wheel-soil interaction model and optimization of robot kinematics.

In this paper, we propose a vision-based simultaneous localization and map-building (SLAM) algorithm using compressed extended Kalman filter (CEKF). SLAM addresses the problem of locating a mobile robot in unknown environments. Extended Kalman filters (EKF) are widely used to solve this problem. However, this filter is very time consuming. To reduce the computational complexity, we apply a CEKF to stereo images while compensating for some of the limitation shown in previous implementations of CEKF. Moreover, we estimate the full DOF, its position and pose, of the mobile robots which is required when operating in the outdoor environment. Outdoor experiments have been conducted to test the effectiveness of the proposed SLAM algorithm.

This paper introduces a link-type tracked mobile robot developed for demining operations. The robot consists of three parts: a front frame, a rear frame, and a body. The front frame is connected to the rear frame by a rotational passive adaptation mechanism, which is a driving mechanism of the robot. This passive adaptation mechanism enables the proposed robot good adaptability to uneven terrain including stairs. The link structure gives rise to a small, simple, and energy efficient vehicle. In addition, the new demining system adaptable to the proposed tracked mobile robot is discussed on how to clear small anti-personnel mines with a non-explosive method. Both the motion of the demining rake and the design parameters of the demining unit are analyzed. Finally, a presentation follows on how the new demining system can unearth small anti-personnel mines with a non-explosive method.Copyright © 2004 by ASME

The purpose of this project is to develop a mobile robot for hazardous terrain exploration. The exploration of hazardous terrain requires the development of a passive mechanism adaptable to such terrain and a sensing system for obstacle avoidance, as well as a remote control. We designed a new mobile robot, the Ronahz 6-wheel robot, which uses a passive mechanism that can adapt to hazardous terrains and building stairways without any active control. The suggested passive linkage mechanism consists of a simple four-bar linkage mechanism. In addition, we install a stereo vision system for obstacle avoidance, as well as a remote control. Wide dynamic range CCD cameras are used for outdoor navigation. A stereo vision system commonly requires high computational power. Therefore, we use a new high-speed stereo correspondence algorithm, triangulation, and iterative closest point (ICP) registration to reduce computation time. Disparity maps computed by a newly proposed, high-speed method are sent to the operator by a wireless LAN equipment. At the remote control site, a three-dimensional digital map around a mobile robot is built by ICP registration and reconstruction process, and this three-dimensional map is displayed for the operator. This process allows the operator to sense the environment around the robot and to give commands to the mobile robot when the robot is in a remote site.Copyright © 2004 by ASME