Search

Human rescuers, who carry out urban search and rescue (USAR) missions, frequently enter dangerous zones to search for survivors. In these zones, rescuer's life may be threatened. For this reason, rescue robots are expected to become useful work partner for urban search and rescue missions. This paper presents platform design of FUMA, which is a four-wheeled rescue robot with a 1-DOF arm for environment information gathering. Only one arm of FUMA has two benefits for both mobility and teleoperation. Using this arm, FUMA can climb higher obstacles than simple four-wheeled robot. Human rescuers can operate FUMA at a safe distance while the missions are carried out. To teleoperate robot systems efficiently, the key is the interface between the robot systems and operators. During rescue operation, an operator of FUMA usually relies on the images from two fish-eye cameras installed at the top of the arm, as they provide wide images of the environment with the robot included in those views.

Modeling and control of flexible beams has received a great deal of attention in recent years. When a tip-body of a flexible beam is a rigid body, not only bending vibration but also torsional vibration will occur. In this paper, we model the coupled bending and torsional vibrations as a set of decoupled partial differential equations with a set of coupled boundary conditions. The whole set of dynamic equations will be rewritten together as an evolution equation in a properly defined Hilbert space. It will be proven that a differential operator governing the system is selfadjoint and has a compact inverse. A stabilizing feedback control law of a rotation motor will be established on the basis of modal analysis. Experimental methods are discussed in detail, and the results demonstrate the effectiveness of the dynamic model and the proposed control law.

This paper treats the optimal control of temperature distribution by using scanning control. The temperature distribution is governed by a partial differential equation with a forcing term of the Dirac delta function type. The problem is to minimize the deviation of temperature distribution from the desired distribution at a given time by controlling both intensity and position of the heat source. On the basis of a set of ordinary non-linear differential equations derived by the Galerkin approximations, the optimal control is calculated by employing an iterative algorithm. Several numerical results are presented.

When a flexible arm is rotated by a motor about an axis through the arm's fixed end, transverse vibration may occur. The motor torque should be controlled in such a way that the motor rotates by a specified angle, while simultaneously stabilizing vibration of the flexible arm so that it is arrested as soon as possible at the end of rotation. In this paper, we first derive a partial differential equation and a set of boundary conditions governing the vibration. Then, a feedback control system which incorporates a dynamic compensator is designed using sensor outputs. A set of experiments has been constructed to demonstrate control strategies for a flexible arm, where a strain gage was used as a vibration sensor and a microcomputer was equipped as a controller. Several satisfactory experimental results are shown.

A flexible manipulator having a parallel drive mechanism is treated. Only the forearm of the manipulator is assumed to be flexible, because it is slender and carries a heavy end-effector. Both the elastic vibration of the arm, which occurs owing to flexibility, and the position of the driving motors should be controlled simultaneously. A partial differential equation and the boundary conditions that govern the clastic vibration of the arm are derived, and the related eigenvalue problem is solved. On the basis of a finite-dimensional modal model of the distributed-parameter systems, an optimal feedback control system is constructed by using the outputs of several sensors.

A flexible manipulator having a parallel drive mechanism is treated. Only the forearm of the manipulator is assumed to be flexible, because it is slender and carries a heavy end-ef fector. The elastic vibration of the arm, which is due to its flexibility, and the position of the driving motors should be controlled simultaneously. On the basis of the mathematical model developed in our previous paper, we construct an optimal feedback control system by using the outputs of sev eral sensors. A set of experiments for the point-to-point position control was carried out with a microcomputer, which controls the whole servo system and plays the role of dynamic compensator. Several satisfactory experimental results are shown.

Industrial robots have lumped flexibility at joints as well as distributed flexibility due to the elasticity of their arms. Because of the flexibility, undesirable vibration occurs during motion. In this paper, an equivalent spring which represents the flexibility is introduced, and the equations of motion of robot links and equations of vibration of the end-effector are derived by using Lagrangian dynamics. Motor torques should be controlled in such a way that the required response of the motor positions is achieved and at the same time the vibration disappears as quickly as possible. For measuring the vibration, three accelerometers are installed at the tip of the forearm. A robust feedback control law is derived by using the root locus technique. Several successful experimental results are given.

Modeling and control of flexible beams has received a great deal of attention in recent years. When a tip-body of a flexible beam is a rigid body, not only bending vibration but also torsional vibration will occur. In this paper, we model the coupled bending and torsional vibrations as a set of decoupled partial differential equations with a set of coupled boundary conditions. The whole set of dynamic equations will be rewritten together as an evolution equation in a properly defined Hilbert space. It will be proven that a differential operator governing the system is selfadjoint and has a compact inverse. A stabilizing feedback control law of a rotation motor will be established on the basis of modal analysis. Experimental methods are discussed in detail, and the results demonstrate the effectiveness of the dynamic model and the proposed control law.

ロボットには, 関節に集中した柔らかさやアーム材料の弾性による分布的な柔らかさが存在して, そのために振動が発生する. 本文ではこれらの柔らかさを集約した等価的なばねを想定し, Lagrangeの方法を用いて, リンクの運動方程式と先端荷重の振動方程式を導いている. マニピュレータ前腕の先端に3個の加速度計をとりつけて振動を検出し, モータの位置制御と同時に振動を減衰させるようなフィードバック制御則を導いている. 根軌跡法を用いて, この制御則のロバスト性を考察している.実機について行った実験結果は, 理論的に導いた制御則の有効性を示しており, 先端荷重の運動にともなう振動が速かに減衰することを示している.

This paper studies the stability problem for discrete-time systems with interval time-varying delay. By dividing delay interval into two subintervals, a delay-dependent exponential stability criterion is obtained based on Lyapunov stability theory and reciprocally convex lemma. Furthermore, by assuming that the distribution of time-varying delay is known, the difference of Lyapunov functional is allowed to have positive upper bound for the value of time-varying delay in one subinterval, and a new delay distribution dependent stability criterion is obtained. The obtained result is also extended to cope with the robust delay distribution dependent stability problem for uncertain time-varying delay systems. All the obtained criteria are presented in terms of Linear Matrix Inequalities (LMIs). Finally one numerical example is given to show the effectiveness of the proposed method.

We discuss a design of controller for a two-link flexible arm based on a distributed parameter model. Using Lyapunov method, we derive a PDS controller that the system becomes Lyapunov stable and that the system in neighborhood of a desired configuration becomes asymptotically stable. As the controller is derived based on the distributed parameter model, we can avoid the drawbacks resulting from finite dimensional approximation.

We explain bottom-up and top-down approaches to dynamics of hyper-redundant mechanical systems. In the bottom-up approach we consider snake robots as a typical example of the hyper-redundant mechanical systems for extracting the fundamental property and developing the suitable model representation. We propose a control law for avoiding singularity of the snake robot using redundancy. In the top-down approaches we propose two system theoretic approaches to dynamics of hyper-redundant mechanical systems. The first one is focused on the partial periodicity property in the system. The second one is based on the distributed parameter model. We explain the interaction of both bottom-up and top-down approaches for constructing hyper-redundant mechanical systems.

We discuss a force control problem for a constrained one-link flexible arm based on distributed parameter model. In order to solve the force control problem, we propose a simple feedback control law. This control law consists of bending moment at the root of the flexible arm and its derivative, which regulates simultaneously the force and the rotational angle of the motor without the angle and the force information. Then it provides the exponential stability of the closed-loop system by using energy multiplier method. As the control law is derived based on the distributed parameter model, we can avoid the drawbacks resulting from finite dimensional approximation.

Problems can often occur when a legged robot attempts to walk on irregular or damaged terrain, such as in search and rescue missions during natural and man-made disasters. In some cases, the ground beneath the robot will collapse because of the pressure of its weight, causing the machine to lose its foothold and topple over. This is a point to which we as designers must pay careful attention when designing a robot. Thus, in such irregular areas, the robot should walk carefully so as not to collapse its footholds. To attempt to solve this problem, we proposed the “leg-grope walk” method which allows a quadruped robot to avoid stumbling or causing a large collapse of the surrounding area on weak horizontal planes. Specifically, when the robot puts its foot on the ground, it applies some excess force on the ground and confirms whether the foothold is likely to collapse, so as to choose a foothold will not collapse. In this study, we extended this method to weak and irregular slopes, where slippage needs to be considered. A new walking method was designed using a force distribution method. To validate the method, we show simulation results from force distribution and robotic experiments in various environments. These results demonstrate that our method allows a robot to walk carefully without slipping or stumbling, even when its foothold is lost.