Seokmin Hong

A walking pattern generation method of humanoid robot MAHRU-R

This paper proposes an omni-directional walking pattern generation method for a humanoid robot MAHRU-R. To walk stably without falling down, a humanoid robot needs the walking pattern. Our previous walking pattern method generated the walking pattern with linear polynomials of the zero moment point (ZMP). It implemented the simple walking like forward/backward walking, side step walking and turning. However, this method was not sufficient to satisfy the various walking which is combined by forward/backward walking, side step walking and turning. We needed to upgrade the walking pattern generation method to implement an omni-directional walking. We use the linear inverted pendulum model consisted of ZMP and center of mass in order to simplify the computation of walking pattern. The proposed method assumes that the state of the following stride is same to the state of the current stride. Using this assumption of walking pattern, the proposed method generates the stable walking pattern for various walking. And the proposed scheme generates the ZMP trajectory with the quartic polynomials in order to reduce the fluctuation of ZMP trajectory by various walking. To implement the efficient walking pattern, this method proposes three walking modules: periodic step module, transient step module and steady step module. Each step module utilizes weighted least square method with future ZMP position information. The effectiveness of the proposed method is verified by simulations of various walking. And the proposed method is confirmed by the experiment of real humanoid robot MAHRU-R.

Energy harvesting from walking motion of a humanoid robot using a piezoelectric composite

In this paper, we investigate the energy harvesting of a piezoelectric composite in a knee pad during walking motion. We use a humanoid robot as a test bed for the experiments using the knee pad. The humanoid wears the knee pad hosting the piezoelectric composite, and its operating knee motion mimics the walking based on the data captured from a human. The use of the humanoid enables the motion to be completely repeatable. We quantitatively study the energy harvesting by using the repeated motion. An electromechanical model for the piezoelectric energy harvester is used to estimate power transferred to varied load resistances under the repeated knee motion. With a good agreement between the experiments and the model predictions, we demonstrate power harvesting on the order of ten microWatts.

Walking pattern generation method with feedforward and feedback control for humanoid robots

This paper proposes a new walking pattern generation method for humanoid robots. This paper uses the linear inverted pendulum model (LIPM) which is composed of zero moment point (ZMP) and center of mass (CoM). Based on LIPM, the proposed method consists of feedforward control and feedback control for walking pattern generation of humanoid robots. The linear quadratic regulator(LQR) as a feedback controller tracks the desired ZMP according to footprints of humanoid robots and makes poles of LIPM stable. The feedback controller, pole-zero cancelation by series approximation (PZCSA) plays a role of reducing the inherent property of LIPM and approximating the transfer function of the overall system including LIPM and controllers to be unity. The usefulness of the proposed method is verified by simulations such as arbitrary time intervals of support phases, arbitrary desired ZMP position and sudden changed desired ZMP position. And the validity of the proposed method is confirmed by the experiment of a humanoid robot using a joystick.

Proposal of a Single Nano-Magnet Memory Device

We propose a non-volatile memory device using ferromagnetic (FM) contacts fabricated on a channel exhibiting spin-momentum locking observed in diverse materials with spin-orbit coupling like heavy metals and topological insulators. The writing is enabled by the current induced spin–orbit torque, which has been used previously to switch the storage layer of a magnetic tunnel junction (MTJ). The reading is enabled by a relatively lower current-induced spin voltage measurement through the FM contact, which is high or low depending on the magnetization direction for a particular current direction. This new read mechanism significantly reduces the fabrication difficulties compared with MTJ-based designs. Simpler interconnects and control circuits can be used, since both read and write currents share the same path. Our proposal offers on-cell reference voltage generation with a normal metal contacton the channel at thesame position as theFM, which is expected to improve the performance in a large array. The estimated read signal based on available materials is smaller compared with MTJ, but the noise is also expected to be smaller in our metallic device compared with those involving tunnel barriers.

A novel multi-articular leg mechanism for biped robots inspired by bi-articular muscle

This paper proposes a novel multi-articular leg system for biped robot. The structure and actuation scheme of the mechanism are inspired by redundantly-actuated manipulators, of which actuators span over multiple joints like bi-articular muscles. The proposed mechanism for robotic leg contributes to improved capacity in end-effector force exertion, which allows less performance requirements in individual actuators. For the purpose of evaluation, we provide the static torque-force transmission analysis, as well as walking simulation results conducted in the dynamic simulator. The results indicate that the proposed mechanism can benefit from the improved output force capacity.

A Walking Pattern Generation Method for Humanoid Robots Using Least Square Method and Quartic Polynomial

Humanoid robot has been developed for human’s convenience in the human environment. Humanoid robots like ASIMO, HRP, WABIAN and Johnnie were successively developed (Hirai et al., 1998; Yamaguchi et al., 1999; Kajita et al., 2003; Loffler et al, 2003). Researches on humanoid have been done about balancing, walking pattern generation, motion generation, whole body cooperation and so on. In particular, many walking pattern generation methods have been developed to prevent a robot from tipping over. There are generally two groups for the walking pattern generation (Kajita et al., 2003; Hirukawa, 2006). The first group uses forward dynamics with multiple rigid-body (Hirai et al., 1998; Yamaguchi et al., 1999). This group demands precise information of the robot such as mass, inertia, and center of mass(CoM) of each link, and so on. In this group, the footprints of biped robot are changed for keeping the planned configuration of the robot (Hirukawa, 2006). On the other hand, the second group (Huang et al., 2001; Kajita et al., 2003; Loffler et al., 2003; Harada et al., 2004; Zhu et al., 2004; Oh et al., 2006) utilizes limited knowledge such as the total center of mass and total angular momentum. And this group makes use of the inverted pendulum model for the walking pattern generation and changes the configuration of the robot for keeping the planned footprints (Hirukawa, 2006). Our walking pattern method belongs to the second group. The inverted pendulum model is transferred the complex dynamic equation of humanoid robot into second order differential equation with some assumptions. Most researches on walking pattern generation dealt with periodic walking. However it is difficult to implement various gaits by periodic walking generator. Harada et al. introduced the analytical walking pattern method on real-time generation for coping with change of gait. Kajita et al. proposed the omni-direction walking pattern generation method using the preview control. Zhu et al. proposed walking pattern generation method with fixed ZMP and variable ZMP in order to make the biped walking pattern more human-like and more agile. This paper describes the omni-directional walking pattern method for the humanoid robots using the least square method with the quartic polynomials. And we design that the ZMP trajectory has the slope in single support phase. The CoM trajectory becomes effective in regard of velocity compared to the ZMP trajectory