• ETRI Journal
• /
• 제31권5호
• /
• pp.565-575
• /
• 2009

This paper concerns the use of support vector regression (SVR), which is based on the kernel method for learning from examples, in identification of walking robots. To handle complex dynamics in humanoid robot and realize stable walking, this paper develops and implements two types of reference natural motions for a humanoid, namely, walking trajectories on a flat floor and on an ascending slope. Next, SVR is applied to model stable walking motions by considering these actual motions. Three kinds of kernels, namely, linear, polynomial, and radial basis function (RBF), are considered, and the results from these kernels are compared and evaluated. The results show that the SVR approach works well, and SVR with the RBF kernel function provides the best performance. Plus, it can be effectively applied to model and control a practical biped walking robot.

• 제어로봇시스템학회논문지
• /
• 제13권4호
• /
• pp.358-364
• /
• 2007

Intelligent walking modeling of humanoid robot using learning based neuro-fuzzy system is presented in this paper. Walking pattern, trajectory of the zero moment point (ZMP) in a humanoid robot is used as an important criterion for the balance of the walking robots but its complex dynamics makes robot control difficult. In addition, it is difficult to generate stable and natural walking motion for a robot. To handle these difficulties and explain empirical laws of the humanoid robot, we are modeling practical humanoid robot using neuro-fuzzy system based on the two types of natural motions which are walking trajectories on a t1at floor and on an ascent. Learning based neuro-fuzzy system employed has good learning capability and computational performance. The results from neuro-fuzzy system are compared with previous approach.

• Journal of Electrical Engineering and Technology
• /
• 제11권3호
• /
• pp.751-758
• /
• 2016

The modifiable walking pattern generation (MWPG) algorithm can handle dynamic walking commands by changing the walking period, step length, and direction independently. When an infeasible command is given, the algorithm changes the command to a feasible one. After the feasibility of the navigational command is checked, it is translated into the desired center of mass (CM) state. To achieve the desired CM state, a reference CM trajectory is generated using predefined zero moment point (ZMP) functions. Based on the proposed algorithm, various complex walking patterns were generated, including backward and sideways walking. The effectiveness of the patterns was verified in dynamic simulations using the Webots simulator.

• 한국융합학회논문지
• /
• 제11권5호
• /
• pp.175-182
• /
• 2020

계단 보행 시 로봇의 최적 궤도 계산은 유전자 알고리즘과 계산 토크 컨트롤러를 사용하여 수행되었다. 첫째, 생식, 교배, 돌연변이로 이루어진 실시간 유전 알고리즘 (RCGA)을 사용하여 총 에너지 효율이 최소화되었다. 보폭의 시작과 끝, 그리고 조인트, 각도, 각속도 위치 어셈블리 관련 재현성 조건은 선형 제약이다. 다음은 고르지 못한 제약은 코너 스윙 다리와 계단의 외부와의 충돌을 막기 위한 조건, 운동 학적 특이성을 막기 위한 무릎 관절의 조건 및 진행 방향의 안전은 보장되지 않음 이란 조건을 따른다. 마지막으로, 각 관절의 각도 궤도는 염색체를 근사 계수를 가지는 4차 다항식에 의해 정의된다. 이것은 보통 도보를 의미한다. 이 연구에서는 최적의 궤도의 에너지 효율을 7개의 링크로 구성된 7자유도의 2족 로봇을 통한 컴퓨터 시뮬레이션을 통해 분석했다.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제30권11호
• /
• pp.1092-1101
• /
• 2003

대부분의 3차원 캐릭터 애니메이션에서는 모션 캡쳐 장비를 통해서 포착된 동작 데이타를 이용하여 다양한 지형상에서 캐릭터가 걷는 동작을 표현한다. 이러한 동작 포착 데이타는 실제 사람과 같이 움직이는 동작들을 자연스럽게 표현할 수 있으나, 만약 다양한 지형에 대한 움직이는 동작이 표현할 경우, 지형의 유형에 따라 모든 동작을 캡쳐하여야 하고, 얻어진 동작 데이타를 다른 유형의 캐릭터에 적용할 경우 동작 데이타를 다시 얻거나 기존 동작 데이타를 재편집해야 하는 어려움이 있다. 따라서 본 연구에서는 적은 매개변수들을 사용하여 평지면, 경사면, 계단면 그리고 굴곡면 등 다양한 지형에서의 적응적인 걷는 동작을 생성하기 위한 방법과 골반과 이동하는 다리의 움직임 제적을 산출하는 방법을 제안한다. 이 방법에서는 캐릭터의 신장이나 걷는 속도, 걸음폭 등의 매개변수들을 조절하여 다양한 걸음걸이를 생성할 수 있으며 역운동학(Inverse Kinematics) 개념을 적용하여 관절들의 위치나 각도를 산출하고 관절의 이동 궤적을 계산하기 위해 큐빅 스플라인 곡선을 활용한다.

• 제어로봇시스템학회논문지
• /
• 제12권4호
• /
• pp.315-319
• /
• 2006

Support vector machines in biped humanoid robot are presented in this paper. The trajectory of the ZMP in biped walking robot poses an important criterion for the balance of the walking robots but complex dynamics involved make robot control difficult. We are establishing empirical relationships based on the dynamic stability of motion using SVMs. SVMs and kernel method have become very popular method for learning from examples. We applied SVM to model the practical humanoid robot. Three kinds of kernels are employed also and each result has been compared. As a result, SVM based on kernel method have been found to work well. Especially SVM with RBF kernel function provides the best results. The simulation results show that the generated ZMP from the SVM can be improve the stability of the biped walking robot and it can be effectively used to model and control practical biped walking robot.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년 학술대회 논문집 정보 및 제어부문
• /
• pp.539-541
• /
• 2006

Biped locomotion is a popular research area in robotics due to the high adaptability of a walking robot in an unstructured environment. When attempting to automate the motion planning process for a biped walking robot, one of the main issues is assurance of dynamic stability of motion. This can be categorized into three general groups: body stability, body path stability, and gait stability. A zero moment point (ZMP), a point where the total forces and moments acting on the robot are zero, is usually employed as a basic component for dynamically stable motion. In this rarer, learning based neuro-fuzzy systems have been developed and applied to model ZMP trajectory of a biped walking robot. As a result, we can provide more improved insight into physical walking mechanisms.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.2350-2355
• /
• 2005

In this paper, we introduce biped walking robot which can static walking with 22 degree-of-freedoms. The developed biped walking robot is 480mm tall and 2500g, and 22 RC servo motors are used to actuate. Before made an active algorithm, we generated the motions of robot with the motion simulator which developed using by C language. The two dimension simulator is Based on the inverse kinematics and D-H transform. The simulator implements various motions as inputted the ankle's trajectory. Also we developed a simulator which is applied the principle of inverted pendulum to acquires the center of gravity. As we use this simulator, we can get the best appropriate angle of ankle and pelvis when the robot lifts up its one side leg during the working. We implement the walking motions which is based on the data(angle) getting from both of simulators. The robot can be controlled by text shaped command through RF signal of wireless modem which connected with laptop computer by serial cable.

• 대한기계학회논문집A
• /
• 제24권5호
• /
• pp.1155-1165
• /
• 2000

Due to the irregularity of walking ground and the inaccuracy in trajectory control of a leg, the mechanical shock and slip on the ground can be caused in the landing and supporting legs of a walkin g robot, and the robot may lose walking stability. Especially in a jointed-leg type walking robot, those problems are much more severe than in the pantograph type since the leg-weight of the jointed-leg type walking robot is relatively heavier than that of the pantograph type in general. In order to secure the walking stability for the jointed-leg type quadrupedal robot under development in KIST(Korea Institute of Science and Technology), a balancing algorithm consisting of the leg compliance control and the body posture control is implemented in this paper, and the effectiveness of the algorithm is verified through experiments.

• 전기학회논문지
• /
• 제67권7호
• /
• pp.903-909
• /
• 2018

We propose laser scanner sensor system based walking pattern and tracking method of multiple human. This system uses laser scanners sensors and is applicable to wide and crowded area such as hospital and medical care center. The primary objective of this research is to promote the development of robust, repeatable and transferable software for security system that can automatically detect, track and follow people in public area. We developed the method of human identification for this system. Our method is following: 1. Best-walking pattern data are obtained by the help of human position and direction data obtained by laser scanners. 2. Human identification is conducted by calculating the correlation between the step length of walking human. It becomes possible to conduct human identification even in crowded scenes by estimating the movements of waling human' feet are periodic. In the experiment in the station, some effectiveness of this method became clear.