Search

Many studies are being conducted for the design of autonomous mobile robots which can learn adaptive motions without specific prior knowledge in dynamically changing environments. To this end, it is important to learn general motion rules, while excluding the assumptions limited to specific environments. In this paper, an adaptive motion learning architecture is presented. The architecture enables robots to avoid obstacles in unexperienced environments using the previously entered basic motions and generate generalized motion rules using the learned information. To collect learning data, robot conducts wandering motions to obtain and transmit the sensor information, distance to obstacles, and collision information to the learning architecture, which generates the motion of robot using the basic and learned adaptive motion rules. Lastly, the applicability is verified by learning obstacle avoiding motion in virtual space similar to actual space, by simulation.

Recently, interest of general public toward Ubiquitous Home Network has immensely increased and wireless PAN (Personal Area Network) has attracted strong attentions as short-distance networking solution. Convenience of Wireless PAN technology has attracted more attentions over traditional wired home network devices such as Ethernet, PLC and HomePNA since it requires no cabling work. In the future, home server will be combined with robot to provide functionalities identical to current home service robot as well as to implement more effective and spontaneous server. In this paper, embedded board has proposed as a home server for an efficient control of internal information and conditions of house from remote location and virtual home robot server to be implemented with Zigbee sensor network.

Recently, design semiosis stands out in been developing industrial goods. It is that form of product has meaning. Is changing to company's image with concept of CIPD (Corporate Identity through Product Design). Now, it is that consumer's awareness changes to side that the much wealths prefer product that have better design than product that have function. Accordingly, each company is changing to development of design putting first in product development of function putting first. Is changing by entertainment robot and so on in industrial robot for production as well as robot development purpose. Therefore, need design method of robot based on design to developed robot forward. For this, must decide basis Design from development step of robot. And need correct synthesize use subject and purpose. In this PAPER, recognize about effect of when design compares applied case and other that is not so and applied design to robot. Establish target to apply in design of small size mobile robot and consider above thing.

This paper proposes a simple and effective correction method for the distortion of the picture which came from the distortion of the spherical lens and the state that real image plane and picture image plane are not horizontal. Even though there are many algorithms that corrects for the distortion of the lens, most of the algorithms are modeling lens, take a method to find a proper parameters. This demands time delay depends on the degree of modeling and the amount of calculation. In addition, most of the algorithms is processed with condition that lens is fixed in center of image and a picture image plane and a real image plane are horizontal. Thus, in a system which needs real time image processing such as movable robots, more effective and simpler algorithm is necessary because the amount of the calculation becomes problematic in system. Furthermore, when a robot in motion intent to get data on real image place, where picture plane and real image plane are not horizontal it could be a problem. Therefore, when the distortion of image which occurred from the distortion of the spherical lens, real image plane and picture plane are not horizontal, this paper suggests more effective and simpler algorithms, which corrects for the distortion of the lens with kinematics without modeling

Till now, robot soccer system has much developed. But in the robot soccer system, image system is still one of important devices to recognize the location of robot and color patch is immeasurably important to recognize image. Most teams are using quadrilateral Color patch and oblique color patch, and are making experiments with various IDs in order to overcome their shortcomings, but they tend to return to quadrilateral color patch and oblique color patch in the end. Therefore, in this paper, we proposes a new form of Color patch to promote the merits of existing color patches and to complement their shortcomings in the treatment of image based on the past play and research experiences, applying it to the robot soccer system and thus verifying its performance.