• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.4
• /
• pp.469-475
• /
• 2010

This paper presents controller design of a two wheeled mobile inverted pendulum robot for one man transportation vehicle. Since the overall mass is varying with different drivers, suitable controller gains are obtained through experimental studies. Variation of the center of gravity due to different masses also affects stable balancing control. Thus, the desired balancing angle si required to be modified with respect to different masses. To measure masses for different drivers, a weight scale is used and those data are used for balancing control through communication. The gain scheduling method of using data obtained from experimental studies allows the robot to have stable balancing performances.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.24 no.4
• /
• pp.299-309
• /
• 2014

When biped robots make turns, the ability to walk stably and precisely along any circular path is crucial. In this context, inverse kinematics solutions are found for accurate gait realization, and new zero moment point(ZMP) equations are derived with respect to the cyclindrical coordinate system to facilitate generation of stable walking patterns. Then, appropriate steady and transitional walking patterns are both proposed in form of time functons. Subsequently, walking patterns for a path but of different speeds are generated using the functions and associated formulas, and preliminarily checked for stability based on the ZMP equations. Upon comparison of those cases, one can see how and when robots may fall down during circular walking. Finally, those patterns are put to test on the sample robot by ADAMS(R) along with the inverse kinematics solutions and a new balance control scheme compensating for insufficient stability particulary during the initial transition period. Test results show that the robot can walk along the circular path as predicted at a resonably high speed despite the distributed mass and ground contact effects, validating effectiveness of the suggested approach.

• Journal of Mechanical Science and Technology
• /
• v.20 no.5
• /
• pp.612-620
• /
• 2006

This paper proposes an optimal trajectory generation method for biped robots for walking up-and-down stairs using a Real-Coded Genetic Algorithm (RCGA). The RCGA is most effective in minimizing the total consumption energy of a multi-dof biped robot. Each joint angle trajectory is defined as a 4-th order polynomial of which the coefficients are chromosomes or design variables to approximate the walking gait. Constraints are divided into equalities and inequalities. First, equality constraints consist of initial conditions and repeatability conditions with respect to each joint angle and angular velocity at the start and end of a stride period. Next, inequality constraints include collision prevention conditions of a swing leg, singular prevention conditions, and stability conditions. The effectiveness of the proposed optimal trajectory is shown in computer simulations with a 6-dof biped robot model that consists of seven links in the sagittal plane. The optimal trajectory is more efficient than that generated by the Modified Gravity-Compensated Inverted Pendulum Mode (MGCIPM). And various trajectories generated by the proposed GA method are analyzed from the viewpoint of the consumption energy: walking on even ground, ascending stairs, and descending stairs.

• Proceedings of the KIEE Conference
• /
• 2006.07d
• /
• pp.1937-1938
• /
• 2006

최근 들어 안정적인 보행 패턴 생성을 위해서 많은 방법들이 제안 되고 있다. 대부분의 논문에서 주기적인 보행에 대한 연구는 이루어지고 있으나 첫 보행 구간과 마지막 보행 구간에 대한 분석은 이루어지지 않고 있다. 본 논문은 첫 보행 구간과 마지막 보행 구간에 대한 분석을 통해 기존의 역 진자 모델(Inverted pendulum model)을 기반으로 부드러운 무게 중심의 궤적을 생성하는 해석적 방법을 제안한다. 이를 위해 먼저 정현파 함수를 이용해 영 모멘트 위치(ZMP, Zero Moment Point) 궤적을 설계한다. 영 모멘트 위치 궤적 설계 시 첫 보행 구간과 마지막 보행 구간에 대해 영 모멘트 위치와 무게 중심 간의 비 최소 위상(non-minimum phase) 시스템의 특성을 이용한다. 제안된 방법을 이용하여 주기적인 보행 구간 및 첫 보행 구간과 마지막 보행 구간에서 부드러운 무게 중심 궤적이 생성됨을 시뮬레이션을 통해 구현하여 제안된 방법의 유효성을 보인다.

• Journal of Institute of Control, Robotics and Systems
• /
• v.10 no.10
• /
• pp.922-929
• /
• 2004

In this paper, we develop control hardware such as an FPGA based general purposed intelligent controller with a DSP board to solve nonlinear system control problems. PID control algorithms are implemented in an FPGA and neural network control algorithms are implemented in a BSP board. An FPGA was programmed with VHDL to achieve high performance and flexibility. The additional hardware such as an encoder counter and a PWM generator can be implemented in a single FPGA device. As a result, the noise and power dissipation problems can be minimized and the cost effectiveness can be achieved. To show the performance of the developed controller, it was tested fur nonlinear systems such as a robot hand and an inverted pendulum.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2011.01a
• /
• pp.279-280
• /
• 2011

2바퀴이상의 로봇은 중심점을 기준으로 안정화가 이루어진다. 그러나 2바퀴이하의 로봇으로 수직 자세를 유지하기 위해서는 로봇자체를 기울여 중심점을 이동하므로 수평을 유지할 수 있다. 그러나 이러한 중심점의 이동은 속도나 방향성분이 같이 출력되므로 정확한 센서의 계산이 요구되고 정밀한 제어를 필요로 한다. 또한 많은 구조물로 인해 장애물 인식 및 자율주행 알고리즘 등이 필요하며 장시간 정보획득과 무인기 연동을 위한 빠른 움직임을 가져야한다. 위의 2조건을 만족하기 위한 구성으로 최근들어 두 바퀴를 가지는 모바일 역진자 로봇에 대한 연구가 활발히 이루어지고 있다. 이는 서비스 및 주행 로봇의 알고리즘이 휴머노이드에서 모바일 역진자 로봇으로 변화되었기 때문이다. 모바일 역진자 로봇은 휴머노이드에 비하여 사용되는 모터의 수가 적고 균형을 잡으려면 관절마다 값비싼 고성능 모터가 필요하며 이를 가동하려면 전력도 많이 소모되며 대용량 배터리를 장착할 수밖에 없게 된다. 반면 바퀴로 움직이는 로봇은 전력이 적게 들고 이동도 쉽다. 따라서 본 연구에서는 IMU를 이용한 간단하면서도 정확한 센서의 연산 방법과 이를 이용한 자세제어 방법을 연구한다.

• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.11
• /
• pp.912-918
• /
• 2016

This paper presents various strategies for humanoid vehicle driving and egress tasks. For driving, a tele-operating system that controls a robot based on a human operator's commands is built. In addition, an autonomous assistant module is developed for the operator. Normal position control can result in severe damage to robots when they egress from vehicles. To prevent this problem, another approach that mixes various joint control techniques is adopted in this study. Additionally, a footplate is newly designed and attached to the vehicle floor for the ground landing phase of the egress task. The attached plate enables the robot to step down onto the ground in a safe manner. For stable locomotion, a balance controller is designed for the humanoid. For the design of the controller, the robot is modeled using an inverted pendulum that consists of a spring and a damper. Then, a state feedback controller (with pole placement and a state observer) is built based on the simplified model. Many approaches that are presented in this paper were successfully applied to a full-sized humanoid, DRC-HUBO+, in the DARPA Robotics Challenge Finals, which were held in the United States in 2015.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.10
• /
• pp.1029-1036
• /
• 2011

An energy-efficient reference walking trajectory generation algorithm is suggested utilizing allowable ZMP (Zero-Moment-Point) region, which maxmizes the energy efficiency for cyclic gaits, based on three-dimensional LIPM (Linear Inverted Pendulum Model) for biped robots. As observed in natural human walking, variable ZMP manipulation is suggested, in which ZMP moves within the allowable region to reduce the joint stress (i.e., rapid acceleration and deceleration of body), and hence to reduce the consumed energy. In addition, opimization of footstep planning is conducted to decide the optimal step-length and body height for a given forward mean velocity to minimize a suitable energy performance - amount of energy required to carry a unit weight a unit distance. In this planning, in order to ensure physically realizable walking trajectory, we also considered geometrical constraints, ZMP stability condition, friction constraint, and yawing moment constraint. Simulations are performed with a 12-DOF 3D biped robot model to verify the effectiveness of the proposed method.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.7 no.4
• /
• pp.65-73
• /
• 1997

This paper proposes a new method of the structure pruning of dynamic recurrent neural networks (DRNN) using evolutionary computations. In general, evolutionary computations are population-based search methods, therefore it is very useful when several different properties of neural networks need to be optimized. In order to prune the structure of the DRNN in this paper, we used the evolutionary programming that searches the structure and weight of the DRNN and evolution strategies which train the weight of neuron and pruned the net structure. An addition or elimination of the hidden-layer's node of the DRNN is decided by mutation probability. Its strategy is as follows, the node which has mhnimum sum of input weights is eliminated and a node is added by predesignated probability function. In this case, the weight is connected to the other nodes according to the probability in all cases which can in- 11:ract to the other nodes. The proposed pruning scheme is exemplified on the stabilization and position control of the inverted-pendulum system and visual servoing of a robot manipulator and the effc: ctiveness of the proposed method is demonstrated by numerical simulations.