• 제어로봇시스템학회:학술대회논문집
• /
• 1999.10a
• /
• pp.292-297
• /
• 1999

Evolving in artificial agent is an extremely difficult problem, but on the other hand, a challenging task. At present the studies mainly centered on single agent learning problem. In our case, we use simulated soccer to investigate multi-agent cooperative learning. Consider the fundamental differences in learning mechanism, existing reinforcement learning algorithms can be roughly classified into two types-that based on evaluation functions and that of searching policy space directly. Genetic Programming developed from Genetic Algorithms is one of the most well known approaches belonging to the latter. In this paper, we give detailed algorithm description as well as data construction that are necessary for learning single agent strategies at first. In following step moreover, we will extend developed methods into multiple robot domains. game. We investigate and contrast two different methods-simple team learning and sub-group loaming and conclude the paper with some experimental results.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.04a
• /
• pp.103-106
• /
• 2004

본 본문은 하이퍼카오스, 로렌츠, 해밀톤 방정식과 같은 여러 종류의 카오스 회로를 이동 로봇에 내장하여 카오스 이동 로봇을 구성하고 이 카오스 이동 로봇이 어느 임의 평면을 카오스 궤적을 가지고 탐색하다가 목표물에 근접하거나 탐색하고자 하는 목표물이 확인되면 집중적인 탐색을 실행하는 방법을 제시하고 그 결과를 검증하였다. 목표물 탐색에서는 장애물 회피와 유사한 카오스 궤적을 가지고 탐색하도록 하는 알고리즘을 개발하고 그 결과를 검증하였으며 이에 대한 타당성을 확인하였다.

• Proceedings of the KIEE Conference
• /
• summer
• /
• pp.932-934
• /
• 2004

BLDC Motor is used in robots requiring a precisive motion recently. This paper presents the optimal design reducing the rotor inertia in order to improve the driving characteristic of BLDC motor driving robots, The optimal design was performed by using a parallel Genetic Algorithm which is superior at searching objective functions for the comflicated models having several optimal points. Therefore, objective function for optimization is rotor inertia and efficiency.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.1
• /
• pp.130-134
• /
• 2012

In this paper, we propose a space partitioning technique for swarm robots by using the Centroidal Voronoi Tessellation. The proposed method consists of two parts such as space partition and collision avoidance. The space partition for searching a given space is carried out by a density function which is generated by some accidents. The collision avoidance is implemented by the potential field method. Finally, the numerical experiments show the effectiveness and feasibility of the proposed method.

• ETRI Journal
• /
• v.36 no.6
• /
• pp.913-923
• /
• 2014

This paper proposes a global mapping algorithm for multiple robots from an omnidirectional-vision simultaneous localization and mapping (SLAM) approach based on an object extraction method using Lucas-Kanade optical flow motion detection and images obtained through fisheye lenses mounted on robots. The multi-robot mapping algorithm draws a global map by using map data obtained from all of the individual robots. Global mapping takes a long time to process because it exchanges map data from individual robots while searching all areas. An omnidirectional image sensor has many advantages for object detection and mapping because it can measure all information around a robot simultaneously. The process calculations of the correction algorithm are improved over existing methods by correcting only the object's feature points. The proposed algorithm has two steps: first, a local map is created based on an omnidirectional-vision SLAM approach for individual robots. Second, a global map is generated by merging individual maps from multiple robots. The reliability of the proposed mapping algorithm is verified through a comparison of maps based on the proposed algorithm and real maps.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.37 no.3
• /
• pp.27-38
• /
• 2000

This paper proposes how the recurrent neural network controller for a Khepera mobile robot with an obstacle avoiding ability can be determined by co-adaptation of the evolution and learning, The proposed co-adaptation scheme consists of two folds： a population of NN controllers are evolved by the genetic algorithm so that the degree of obstacle avoidance might be reduced through the global searching and each NN controller is trained by CRBP learning so that the running behavior is adapted to its outer environment through the local searching. Experimental results shows that the NN controller coadapted by evolution and learning outperforms its non-learning equivalent evolved by only genetic algorithm in both the ability of obstacle avoidance and the convergence speed reaching to the required running behavior.

• Journal of the Institute of Convergence Signal Processing
• /
• v.5 no.4
• /
• pp.256-262
• /
• 2004

A key component of an autonomous mobile robot is to localize itself and build a map of the environment simultaneously. In this paper, we propose a vision-based localization and mapping algorithm of mobile robot using fisheye lens. To acquire high-level features with scale invariance, a camera with fisheye lens facing toward to ceiling is attached to the robot. These features are used in mP building and localization. As a preprocessing, input image from fisheye lens is calibrated to remove radial distortion and then labeling and convex hull techniques are used to segment ceiling and wall region for the calibrated image. At the initial map building process, features we calculated for each segmented region and stored in map database. Features are continuously calculated for sequential input images and matched to the map. n some features are not matched, those features are added to the map. This map matching and updating process is continued until map building process is finished, Localization is used in map building process and searching the location of the robot on the map. The calculated features at the position of the robot are matched to the existing map to estimate the real position of the robot, and map building database is updated at the same time. By the proposed method, the elapsed time for map building is within 2 minutes for 50㎡ region, the positioning accuracy is ±13cm and the error about the positioning angle of the robot is ±3 degree for localization.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.40 no.4
• /
• pp.723-730
• /
• 2015

In this paper, a location tracking compensation algorithm based on the Least-Squares Method ($LCA_{LSM}$) was proposed to improve the autonomous tracking efficiency for the docent robot in exhibition hall, and the performance of the $LCA_{LSM}$ is analyzed by several practical experiments. The proposed $LCA_{LSM}$ compensates the collected location coordinates for the robot using the Least-Squares Method (LSM) in order to reduce the cumulated errors that occur in the Encoder/Giro sensor (E/G) and to enhance the measured tracking accuracy rates in the autonomous tracking of the robot in exhibition hall. By experiments, it was confirmed that the average error reduction rates of the $LCA_{LSM}$ are higher as 4.85% than that of the $LCA_{KF}$ in Scenario 1 (S1) and Scenario 2 (S2), respectively on the location tracking. In addition, it was also confirmed that the standard deviation in the measured errors of the $LCA_{LSM}$ are much more low and constant compared to that of the E/G sensor and the $LCA_{KF}$ in S1 and S2 respectively. Finally, we see that the suggested $LCA_{LSM}$ can execute more the stabilized location tracking than the E/G sensors and the $LCA_{KF}$ on the straight lines of S1 and S2 for the docent robot.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.6
• /
• pp.2493-2503
• /
• 2018

Passive dynamic walking exhibits humanoid and energy efficient gaits. However, optimal design of passive walker at multi-variable level is not well studied yet. This paper presents a Chaotic Particle Swarm Optimization (CPSO) algorithm and applies it to the optimal design of flexible passive walker. Hip torsional stiffness and damping were incorporated into flexible biped walker, to imitate passive elastic mechanisms utilized in human locomotion. Hybrid dynamics were developed to model passive walking, and period-one gait was gained. The parameters global searching scopes were gained after investigating the influences of structural parameters on passive gait. CPSO were utilized to optimize the flexible passive walker. To improve the performance of PSO, multi-scroll Jerk chaotic system was used to generate pseudorandom sequences, and chaotic disturbance would be triggered if the swarm is trapped into local optimum. The effectiveness of CPSO is verified by comparisons with standard PSO and two typical chaotic PSO methods. Numerical simulations show that better fitness value of optimal design could be gained by CPSO presented. The proposed CPSO would be useful to design biped robot prototype.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.1
• /
• pp.139-146
• /
• 2020

The disadvantage of the general robot cleaner is that it works only on the designated route, so it is impossible to clean the place outside the designated route. Therefore, in this study, the direction control methodology for searching the place other than the designated route based on the finger recognition technology was studied to compensate for the shortcomings of the existing cleaner. Raspberry Pi was used as the main controller and Open CV program was used to recognize the number of fingers. To verify the validity of the proposed methodology, a finger recognition algorithm was implemented using Python language, and as a result of adopting the Logitech C922, the success rate was 100% at 90cm and 70% at 110cm, respectively.