• Journal of the Korea Computer Graphics Society
• /
• v.27 no.1
• /
• pp.9-17
• /
• 2021

In this paper, we present a simple and fast supervised learning framework based on model predictive control so as to learn motion controllers for a physic-based character to track given example motions. The proposed framework is composed of two components: training data generation and offline learning. Given an example motion, the former component stochastically controls the character motion with an optimal controller while repeatedly updating the controller for tracking the example motion through model predictive control over a time window from the current state of the character to a near future state. The repeated update of the optimal controller and the stochastic control make it possible to effectively explore various states that the character may have while mimicking the example motion and collect useful training data for supervised learning. Once all the training data is generated, the latter component normalizes the data to remove the disparity for magnitude and units inherent in the data and trains an artificial neural network with a simple architecture for a controller. The experimental results for walking and running motions demonstrate how effectively and fast the proposed framework produces physics-based motion controllers.

• Journal of Power System Engineering
• /
• v.17 no.3
• /
• pp.82-88
• /
• 2013

For commercialized servo drives of the motion stage to include embedded controller, external terminal is provided for tracking command and encoder output, but internal terminal is not for control input. Thus, it is difficult to combine out signal of embedded controller with that of external compensator such as disturbance observer. In this study, for precise tracking control of motion stage without hardware change of the servo drive, tacking control system is composed of an inner loop of servo drive and an outer loop of disturbance observer. Then, the control system is designed so that the output response of actual plant corresponds with nominal model's in transient state as well as in steady state. Finally, the experiment results show that the designed control system is effective to reconcile actual plant behavior with nominal model under nonlinear friction and parameter perturbation.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.3 no.2
• /
• pp.66-73
• /
• 2008

The pipes have been laid underground while the industry is developing. We have to take maintenance procedure when the pipes are cracked or ruptured. It is very difficult jop to check pipe's crack because the pipes are narrow and laid underground. Using in-pipe robot, we can check the conditions of inner section of pipes, therefore, we designed a crawler type robot to search cracked pipe. In this paper, we have made a special focus on the control of the robot using differential drive algorithm to move in curved section of pipes. The detailed design of the robot with experimental result show the effectiveness of the robot in pipe maintenance.

• Journal of Platform Technology
• /
• v.6 no.3
• /
• pp.23-30
• /
• 2018

A motion pattern generation is a process of calculating a certain stable motion trajectory for stably operating a certain motion. A motion control is to make a posture of a robot stable by eliminating occurring disturbances while a robot is in operation using a pre-generated motion pattern. In this paper, a general method of motion pattern generation for a biped walking robot using universal approximator, learning neural networks, is proposed. Existing techniques are numerical methods using recursive computation and approximating methods which generate an approximation of a motion pattern by simplifying a robot's upper body structure. In near future other approaches for the motion pattern generations will be applied and compared as to be done.

• Journal of the Korean Society of Industry Convergence
• /
• v.18 no.2
• /
• pp.119-128
• /
• 2015

In this paper, we propose a new motion control technology to design robust control system of industrial robot. The system modeling of robotic manipulation tasks with constraints is presented, and the control architecture for unconstrained and constrained motion system with parametric uncertainties is synthesized. The optimal reference of robot manipulator is generated by the reference controller as a discrete state system and the control behavior of constrained system which has poor modeling information and time-invariant constraint function is improved motion control system is successfully evaluated by experiment to the desired tasks.

• Journal of the Korea Computer Graphics Society
• /
• v.12 no.2
• /
• pp.1-7
• /
• 2006

블렌드 쉐입 기반 얼굴 애니메이션을 위해 기저 모델(Expression basis)을 생성하는 방법을 크게 두 가지로 구분하면, 애니메이터가 직접 모델링을 하여 생성하는 방법과 통계적 방법에 기초하여 모델링하는 방법이 있다. 그 중 애니메이터에 의한 수동 모델링 방법으로 생성된 기저 모델은 직관적으로 표정을 인식할 수 있다는 장점으로 인해 전통적인 키프레임 제어가 가능하다. 하지만, 표정 공간(Expression Space)의 일부분만을 커버하기 때문에 모션데이터로부터의 재복원 과정에서 많은 오차를 가지게 된다. 반면, 통계적 방법을 기반으로 한 기저모델 생성 방법은 거의 모든 표정공간을 커버하는 고유 얼굴 모델(Eigen Faces)을 생성하므로 재복원 과정에서 최소의 오차를 가지지만, 시각적으로 직관적이지 않은 표정 모델을 만들어 낸다. 따라서 본 논문에서는 수동으로 생성한 기저모델을 유사 고유 얼굴 모델(Quasi-Eigen Faces)로 변형하는 방법을 제시하고자 한다. 결과로 생성되는 기저 모델은 시각적으로 직관적인 얼굴 표정을 유지하면서도 통계적 방법에 의한 얼굴표정 공간의 커버 영역과 유사하도록 확장할 수 있다.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.9
• /
• pp.1385-1391
• /
• 2017

This work presents a synchronized on-line trajectory generation algorithm for mechanical system with multiple degrees-of-freedom. Proposed algorithm is designed to generate time-optimized trajectories and synchronized trajectories under the constraints such as maximum speed, acceleration, deceleration, non-zero initial velocity, etc. Also, because of small computation time, therefore this can be applied in real-time and it is easier to change the trajectory when an event occurs. We verified the feasibility through various trajectory generation.

• Journal of the Korean Society of Industry Convergence
• /
• v.17 no.2
• /
• pp.35-41
• /
• 2014

This paper describes the obstacle avoidance architecture to walk safely around in factory and home environment, and presents methods for path planning and obstacle avoidance for the humanoid robot. Solving the problem of obstacle avoidance for a humanoid robot in an unstructured environment is a big challenge, because the robot can easily lose its stability or fall down if it hits or steps on an obstacle. We briefly overview the general software architecture composed of perception, short and long term memory, behavior control, and motion control, and emphasize on our methods for obstacle detection by plane extraction, occupancy grid mapping, and path planning. A main technological target is to autonomously explore and wander around in home environments as well as to communicate with humans.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2001.11a
• /
• pp.611-616
• /
• 2001

본 논문에서는 실세계에서의 사용자 동작을 가상 공간의 캐릭터를 통해 나타낼 수 있는 제어 방법에 대해 논한다. 사용자의 캐릭터는 스테레오 카메라 영상으로부터 얻어지는 5개의 3차원 좌표값을 이용하여 표현된다. 영상에서 구하는 3차원 좌표값은 노이즈를 포함하는 불안정한 데이터이므로 자연스럽고 부드럽게 변형할 수 있도록 프레임 내에서 보정 작업이 필요하다. 보정 후 안정된 위치 데이터는 부드럽게 변형할 수 있도록 프레임 내에서 보정 작업이 필요하다. 보정 후 안정된 위치 데이터는 신체 제약조건을 만족하는 범위에서 동역학(inverse kinematic)을 이용하여 관절각 데이터로 변형한다. 하지만 이 방법은 수학적인 계산을 통해 나온 결과이기 때문에 때때로 사용자가 운하지 않는 동작을 만들어 내는 단점을 가지고 있다. 이를 극복하기 위해 관절각 제약조건을 만족하는 데이터만으로 보간 작업을 한다.

• Journal of Korea Multimedia Society
• /
• v.11 no.8
• /
• pp.1138-1145
• /
• 2008

Recently, performance-driven facial animation has been popular in various area. In television or game, it is important to guarantee real-time animation for various characters with different appearances between a performer and a character. In this paper, we present a new facial animation approach based on motion capture. For this purpose, we address three issues: facial expression capture, expression mapping and facial animation. Finally, we show the results of various examination for different types of face models.