Kazuhito Yokoi

Introduction to Humanoid Robotics

This book is for researchers, engineers, and students who are willing to understand how humanoid robots move and be controlled. The book starts with an overview of the humanoid robotics research history and state of the art. Then it explains the required mathematics and physics such as kinematics of multi-body system, Zero-Moment Point (ZMP) and its relationship with body motion. Biped walking control is discussed in depth, since it is one of the main interests of humanoid robotics. Various topics of the whole body motion generation are also discussed. Finally multi-body dynamics is presented to simulate the complete dynamic behavior of a humanoid robot. Throughout the book, Matlab codes are shown to test the algorithms and to help the readers understanding.

A switching command-based whole-body operation method for humanoid robots

This paper introduces a switching command-based whole-body operation method for humanoid robots. Humanoid robots are biped machines possessing multiple degrees of freedom (DOF). Due to the complexity of their multi-DOF structure, and the difficulty in maintaining postural stability, whole-body operation of humanoid robots is fundamentally different from traditional fixed-base manipulators or stable-base mobile manipulators. By studying the shifts in locus of attention between human body joints during task execution, we developed a switching command-based operation method that allows the operator to select only the necessary points of the humanoid robots body for manipulation. Whole-body motion satisfying the desired movements of the selected points is generated using an inverse-kinematics motion generation scheme. This switching operation method enables flexible whole-body operation of humanoid robots using simple input devices. The proposed whole-body operation method is implemented as a teleoperation system using two 3-DOF joysticks to operate a 30-DOF humanoid robot (HRP-1S) developed in the Humanoid Robotics Project (HRP) of the Ministry of Economy, Trade, and Industry of Japan. Experiments teleoperating HRP-1S confirmed the effectiveness of our method.

Whole body teleoperation of a humanoid robot - a method of integrating operator's intention and robot's autonomy

This paper proposes a new method for the teleoperation of humanoid robots which integrates human operators intention with robots autonomy. Getting hints from human conscious and subconscious motion generations, we propose a method which generates whole body motions of a humanoid robot satisfying operators desired movement of specific points on which the operator focuses as well as the robots balance. The proposed method is based on a whole body momentum control. We also introduce a function of automatic adjustments of the position of the center of mass and torso orientation in order to expand the reachable area of a humanoid robot. The effectiveness of the method is confirmed by using humanoid robot simulator OpenHRP with physical parameters of real humanoid robot HRP-1S.

Running pattern generation and its evaluation using a realistic humanoid model

This paper describes a possibility of a running humanoid robot based on the physical properties of an existing humanoid robot HRP-2L. To generate running motion, we develop a general method to control the total linear/angular momentum of a multi-link system. By this method, we can generate a reliable running pattern. Conducting simulations which take into account physical restrictions, we show that our humanoid robot can run at least at 0.5 m/s.

Numerical Methods for Reachable Space Generation of Humanoid Robots

In view of the importance of workspace to robotic design, motion planning and control, we study humanoid reachable spaces. Due to the large number of degrees of freedom, the complexity and special characteristics of humanoid robots that conventional robots do not possess, it would be very difficult or impractical to use analytical or geometric methods to analyze and obtain humanoid reachable spaces. In this paper, we develop two numerical approaches — the optimization-based method and the Monte Carlo method — to generate the reachable space of a humanoid robot. We first formulate the basic constraints (including kinematic constraint and balance constraint) that a humanoid robot must satisfy in manipulation tasks. We then use optimization techniques to build mathematical models for boundary points by which the reachable boundary is formed. This method gives rise to an approximation of the reachable space with accurate boundary points. On the other hand, the Monte Carlo method is relatively simple and more suitable for the visualization of robotic workspace. To utilize the numerical results by the Monte Carlo method, we propose an approach to build a database. We present the algorithms with these two methods and provide illustrating examples conducted on the humanoid HRP-2.

Dynamic lifting by whole body motion of humanoid robots

A motion control method of lifting a heavy object up to a higher position with humanoid robots is developed. The key issue of lifting motion is how to reduce the load on humanoid arms in which low-power actuators are implemented. The use of singular postures of arms is well-known to avoid actuator saturation of the arms. By combining two different kinds of humanoid motions such as accelerating an object upward and sliding the body into under the object, we propose a method that enables to transit one singular posture of arms to another while lifting the object. Simulation results show the effectiveness of the proposed method for reducing the load on the arms. We realize a motion of lifting a heavy object dynamically with the humanoid robot HRP-2 through experiment.

On robotic trajectory planning using polynomial interpolations

Trajectory planning is a classic topic in robotics. Although cubic splines are widely used to interpolate the trajectory in the literature, they have drawbacks in satisfying more boundary conditions for high dynamic performance, say, the control of acceleration at the end points. On the other hand, if quartic splines are employed, more conditions are needed to determine the coefficients of the fourth-order polynomials for the trajectory. In this paper, we present a method to interpolate the trajectory by combining third-order and fourth-order polynomials. We use two quartic polynomials for the first and last segments of the trajectory, and cubic ones for other segments. The trajectory can be uniquely determined by a set of path points for boundary conditions including accelerations at end points. We also investigate the planning of optimal trajectories in the case of variant intervals between adjacent path points. In the optimization, we use multiple objectives including accelerations and jerks. How to determine succinctly the polynomial coefficients is described. Examples are provided for illustration, and comparisons with previous planning method are reported. Good results of application in humanoid robot motion planning for obstacle stepping-over are obtained

“Give me the purple ball” - he said to HRP-2 N.14

This paper reports current experiments conducted on HRP-2 based research on robot autonomy. The contribution of the paper is not focused on a specific area but its objective is to highlight the critical issues that had to be solved to allow the humanoid robot HRP-2 to understand and execute the order ldquogive me the purple ballrdquo in an autonomous way. Such an experiment requires: simple object recognition and localization, motion planning and control, natural spoken language supervision, simple action supervisor and control architecture.

Enhanced Teleoperation Through Virtual Reality Techniques

Teleoperation effectiveness was almost stagnant during several decades since the discovery of the master-slave structure by Goertz in the 50s; in spite of some technical improvements and computer developments [1]. Recent advances on virtual reality (VR) techniques can be considered as the additional ingredient to the total telerobotic renewal, and novel architectures could be developed. Its efficiency is reaching now a considerable level of achievements. It allows a clear improvement in traditional applications such as nuclear or space activities and it also opens many new fields of application, where teleoperation represents a major percentage of robotic development. Build on previous state-of-the-art reports [2, 3, 4], this chapter reviews up-to-date achievements applying virtual reality techniques to teleoperation: we recall some relevant achievements of VR in solving difficult teleoperation problems such as time delay, operator assistance and sharing robot autonomy by combining different supervision strategies or allowing new human-centred teleoperation schemes. The chapter also discusses new robotic applications that have currently appeared which require additional research efforts and call additional investigations on virtual reality techniques. Among them we can find micro- and nano-teleoperation eventually of livings such us cells and DNA molecules, teleoperation of humanoids and animaloids, teleoperation of unmanned air or terrestrial vehicles, hi-fidelity telepresence, multi-operator multirobots teleoperation, etc. These are exemplified along with their specific challenges.

Slip-Turn for Biped Robots

This paper presents a method for generating a turning motion in a humanoid robot by allowing the feet of the robot to slip on the ground. As humans, we exploit the fact that our feet can slip on the ground, and allowing humanoid robots to realize this same motion is a worthwhile study. In this paper, we propose the hypothesis that a turning motion is caused by the effect of minimizing the power generated by floor friction. A model of rotation from this friction is then described based on our hypothesis. The proposed model suggests that only the trajectory and shape of the robots feet determine the amount of rotation from a slip, and that the friction coefficient between the floor and the soles of the robot, as well as the velocity of the feet, do not affect the resultant angle. Verification is conducted through an experiment with a humanoid robot known as HRP-2. Next, to obtain the foot motion necessary to realize the desired rotational angle, the inverse problem is solved by confining the trajectory of the center of the foot to an arc shape. This solution is verified through an experiment with another humanoid robot, HRP-4C.