The momentum equilibrium principle reveals the relative character of the spatial momentum equation in floating-base robotics. This work clarifies how to take advantage of this character in motion generation and the design of a multitask controller for a humanoid robot. We focus especially on the angular momentum of the robot and the inherent redundancy resolution problem referred to as the “momentum distribution problem.” It is shown that with a proper momentum distribution, emergent behaviors (motion synergies) can be obtained that resemble those used by humans. A real-time controller for position- and torque-controlled humanoid robots is proposed, which has a simple structure. The performance of the controller is confirmed in a simulated environment with a number of tasks, including reactive motion control to accommodate external disturbances (both continuous-force and impacts), acrobatic tasks such as somersaults and jumps, dynamic walking with various gaits and variable center of mass height, blind stepping on an unknown obstacle, and balancing on a wobble board under external disturbances.

Contact destabilization after an impact that occurs at high-speed, e.g. when a robot steps on an obstacle of unknown height, can be tackled by injecting angular momentum damping for a short time interval immediately after the impact. This is done by making use of the motion from within the reaction null-space (RNS). The angular momentum damping results in an appropriate arm motion that stabilizes the contacts. An impact at high-speed occurs when the stepping time is very short. In this case, conventional controllers cannot handle the reaction stemming from the swing leg dynamics. A general whole-body controller is designed that makes use of the relative angular acceleration control component to inject the angular momentum damping. The proposed control method is robust; it can deal with obstacles of various height and inclination without altering the feedback gains. The controller is fast since iterative optimization is avoided. The performance is examined via a simulated dynamic stepping.

This work describes how to use reactionless motion control with a free-floating space robot, suggesting thereby some tasks of practical importance. We show that the reactionless motion capability is directly affected by the kinematic structure of the manipulator, depending thereby upon the existence of kinematic redundancy, a typical lower/upper arm subchain and joint offsets. We investigate a seven-DoF redundant manipulator comprising these features and show that approximate reactionless motions can be obtained with the elbow joint only and/or the wrist joints. Using these reactionless motions, we propose three practical maneuvers for eye-in-hand type inspection, arm deploying/stowing and point-to-point motions with partial reactionless motion. Feasibility is verified via numerical simulations.

This paper addresses velocity-level redundancy resolution for S-R-S type redundant manipulators, aiming at precise path following in the presence of kinematic singularities. A novel joint-velocity decomposition scheme that complements and avoids some drawbacks of a previous study based on the Singularity-Consistent method is introduced. As a result, it becomes possible to follow almost any singular path in workspace, including paths passing through bifurcations (nondegenerate singular paths). Further on, the algorithmic singularity introduced in the previous study (the “inner obstacle”) disappears in the new formulation. The problem of wrist singularity is also addressed and solved via a proper self motion component. Only a limited subset of singular paths cannot be handled with the method: those that pass through the double elbow-wrist singularity (an isolated point at the workspace boundary). The effectiveness of the proposed method is illustrated via numerical simulations.

This paper presents a knowledge model of everyday objects for semantic grasp. This model is intended for extracting the grasp areas of everyday objects and approach directions for grasping when the 3D point cloud data and the intended purpose are given. Parts that make up everyday objects have functions related to their manipulation. We therefore represent everyday objects in terms of connected parts of functional units. This knowledge model describes the structure of everyday objects and information on their manipulation. The structure of an everyday object describes component parts of the object in terms of simple shape primitives to provide geometrical information and describes connections between parts with kinematic attributes. The information on the structure is used to map the manipulation knowledge onto the 3D point cloud data. The manipulation knowledge of the object includes the grasp areas and approach directions for the intended purpose. Fine grasps suitable for the intended task can be generated by performing a grasp planning with consideration for stable grasp and the kinematics of the robot in the grasp areas and approach directions.

We examine postural reaction and balance recovery patterns occurring when a standing upright human is subjected to a sudden disturbance within the frontal plane, with the aim of developing balance control strategies for humanoid robots. Five patterns are identified and related to the magnitude of the disturbance. Three of the patterns are modeled and implemented with a small humanoid robot HOAP-2. The models are based on inverted-pendulum and double-pendulum equations, in combination with variable stiffness/damping elements for ensuring appropriate reactions and balance recovery patterns. Supporting foot reaction is minimized within the reaction null space formulation. Special attention is paid to the transitions between the reaction patterns. The experimental data show that the models and the respective controllers can ensure smooth reaction control under both impact-force and continuous-force disturbances.

It is shown that the Reaction Null Space formulation — a method developed for motion analysis and reactionless motion generation of free-floating and flexible-base robots — can be used to fully decouple the end-link dynamics of a kinematically redundant fixed-base robot. Decoupling is achieved thereby via the pseudoinverse of a coupling-inertia submatrix, in quite a different way from the dynamic decoupling achieved via the inertia-weighted generalized inverse of the manipulator Jacobian and known from the Operational Space formulation. The properties of the new formulation are clarified with the help of a comparative study on a representative motion/force control scenario. The simulation results show that the two formulations deliver identical results as far as end-link dynamics are concerned. The new method has an advantage with regard to joint space dynamics, though, which becomes apparent especially in the neighborhood of kinematic singularities, where the inertiaweighted generalized inverse of the manipulator Jacobian is ill-behaved.

We present an integrated approach allowing a free-standing humanoid robot to acquire new motor skills by kinesthetic teaching. The proposed method controls simultaneously the upper and lower body of the robot with different control strategies. Imitation learning is used for training the upper body of the humanoid robot via kinesthetic teaching, while at the same time Reaction Null Space method is used for keeping the balance of the robot. During demonstration, a force/torque sensor is used to record the exerted forces, and during reproduction, we use a hybrid position/force controller to apply the learned trajectories in terms of positions and forces to the end effector. The proposed method is tested on a 25-DOF Fujitsu HOAP-2 humanoid robot with a surface cleaning task.

The task of cleaning a vertical flat surface with the help of a standing humanoid robot is considered. A trajectory/force tracking controller is introduced that is implemented under a mixed position/torque control mode. The arm joints are controlled with a conventional PD controller working under position control, while the ankle joints are torque controlled. The desired force is realized via a force tracking controller using CoM and ZMP position readings obtained from the pressure sensors in the feet. The trajectory/force tracking controller is implemented and experimentally evaluated with a miniature humanoid robot HOAP-2.

High-speed path tracking with a kinematically redundant manipulator mounted on a flexible base is addressed. Thereby, possible vibrations of the base are to be suppressed. In general, the presence of kinematic redundancy allows these two subtasks to be performed simultaneously. In practice, however, this can be done only within very limited areas of workspace, separated by singularity loci that change dynamically while the end-effector tracks the desired path. To avoid controller performance degradation in the neighborhood of such dynamic singularities, and to allow transitions between the distinct workspace areas through singularity boundaries, we propose here a new method for reactionless motion generation within a specified neighborhood of the singularity. In contrast to previous works, this method makes use of a nonzero coupling momentum which is conserved upon entering the neighborhood.

Our ultimate goal is introducing energy-efficient walking patterns to actual humanoid robots. It is known that limit cycle based walking methods, e.g. Passive Dynamic Walking, have such a desired property. Unfortunately, the application of the methods has been limited to simple planar biped models. In this paper, we propose a way of extending limit cycle based walking pattern generation toward a 7DOF 3D biped with ankles, knees, an upper body and with flat feet. This is achieved via first decoupling roll and pitch motions in the frontal/sagittal planes, and then, by designing a limit cycle based walking pattern in the sagittal plane for a planar 5DOF model with ankles, knees and torso. Robustness of the motion is ensured via a feedback control method based on mechanical energy. Then, the planar motion pattern is projected back into 3D space by incorporating dynamic components for gravity compensation and designing a proper trajectory for ankle roll motion. The performances of the walking pattern generator and the controller are confirmed via numerical simulations. The results are presented also as animated motion of a 7DOF 3D biped in the accompanying video.

This paper presents a grasping system for picking up an indicated object in a complex real-world environment using a parallel jaw gripper. The proposed grasping scheme comprises the following three main steps: (1) A user indicates a target object and provides the system with a task instruction on how to grasp it, (2) the system acquires geometric information about the target object and constructs a 3D environment model around the target by stereo vision using the information obtained from the task instruction, and (3) the system finds a grasp point based on grasp evaluation using the acquired information. As an example of the scheme, we examined the picking up of a cylindrical object by grasping at the brim. An important and advantageous feature of this scheme is that the user can easily instruct the robot on how to perform the object-picking task through simple clicking operations, and the robot can execute the task without exact models of the target object and the environment being available in advance.

We aim at experimental verification of the efficiency of natural-motion path tracking (i.e. tracking speed in proportion to the determinant of the Jacobian) in comparison to constant-speed path tracking. This is done first via simulations, with a simple planar manipulator and then with a six-DOF manipulator. From the results it becomes apparent that natural-motion path tracking outperforms constant-speed path tracking in terms of peak joint speed, peak joint torque and total mechanical power, ensuring thereby a higher average tracking speed. The results are also confirmed via experiments with a real six-DOF robotic limb.

This paper describes 3D biped walking generation and control based on Limit Cycle Walking. In our study, we use the simplest possible 3D biped model with three DOFs, incorporating roll and pitch motions in the frontal/sagittal planes, respectively. Our approach dynamically decouples these two motions, stabilizes pitch motion in the sagittal plane via the Limit Cycle Walking approach, introduces robustness for this motion using energy feedback control, and robustness for roll motion based on reference trajectory feedback tracking control. The roll reference trajectory is generated via analysis of the impact dynamics. Performance is verified via simulations.

The capabilities of DLR's multi-DOF humanoid robot Justin are extended with the help of a dynamic torque control component for base reaction minimization. Since the mobile base of the robot comprises springs, reactions induced by arm/torso motions lead to vibrations and deteriorate the performance. The control component is derived from the equation of motion of the robot, represented as an underactuated system, and partitioned into a ¿driven¿ subsystem (one of the arms), and a ¿compensating¿ subsystem (the other arm, with or w/o torso contribution). The control component is then embedded into the existing sophisticated controller structure of Justin, as a feedforward component, with additional control signals from an augmented PD feedback controller. It was possible to obtain satisfactory performance with a very ¿soft¿ compensatory subsystem. The experimental results confirmed the potential of this model-based approach for use in a complex multi-DOF system. As far as we know, this is the first time that a dynamic-coupling compensating controller is applied to a real system of such complexity, utilizing thereby a torque control interface.

With aging, human skin develops a dry condition called senile xerosis. The skin lesion can be prevented by daily skin care such as applying an ointment containing moisturizing factors several times a day. Aged persons, however, have difficulties in accessing the back and rely therefore on nursing care for such treatment. Unfortunately, in underpopulated areas such nursing care may not always be available. To tackle this problem, the concept of a skincare robot is proposed. The feasibility of the concept is then confirmed by designing a real skincare robot and by performing experiments for applying ointment on human's back. The robot developed is able to recognize the shape of the body, to plan the appropriate motion paths for the hand, and to execute the task without applying any excessive forces.

A humanoid robot should be able to keep balance even in the presence of disturbing forces. Studies of human body reaction patterns to sudden external forces (impacts) are useful in developing balance control strategies. In this paper we show how to implement two such reaction patterns, called ankle and hip strategy, using a small humanoid robot. Simple dynamical models in the sagittal plane are employed. The decision for invoking one of the reaction patterns is based on acceleration data measured during the impact. The experiments confirm that the robot is able to react swiftly, resembling the reaction patterns of humans.

This paper describes an experimental system for the teleoperation of a redundant manipulator mounted on a flexible base. Kinematic redundancy is resolved with the help of an additional constraint, obtained from vibration dynamics. The problem of kinematic and algorithmic singularities is addressed via the singularity-consistent method developed in our previous research. Experimental data shows vibration suppression with high efficiency. When no vibrations are present, our approach ensures effective reactionless motion. The stability of the system under teleoperation and while moving around algorithmic and kinematic singularities is also demonstrated

A humanoid robot should be able to keep its balance even in the presence of disturbing forces. Studies of human body reaction patterns to sudden external forces are useful to develop balance control strategies. In this paper we show that two such reaction patterns, called "hip strategy" and "slipping" respectively, can be modeled by means of the reaction null-space method developed earlier for space robots. We use a simple simulation model to confirm the validity. Also, experimental data from a small humanoid robot (HOAP-2) is presented

This paper addresses the problem of naturally looking and energy efficient walk of biped humanoids. We presuppose that such walk requires motion control capability around kinematic singularities, such that the knee can be fully extended. This problem is tackled by adopting the singularity-consistent method developed for manipulator motion control at and around kinematic singularities. We implemented the method with a HOAP-2 humanoid robot, demonstrating stable static walk as a first step in this direction

A kinematic redundancy resolution approach for the S-R-S manipulator is introduced which uses arm plane orientation as the redundancy resolution criterion. The approach can handle both kinematic and algorithmic singularities in a consistent way, without introducing motion instabilities, yielding thereby motions that can be predicted easily by a human operator. The method is verified by experiments with a seven-DOF Mitsubishi Heavy Industries PA10-7C arm and data from both simulations and experiments is presented.

The SC method is applied to a 7-DOF manipulator arm with a spherical wrist and zero joint offsets, such as the Mitsubishi Heavy Industries PA-10 arm. We obtain two vector fields for driving the positioning subchain and one vector field for the wrist, in analytical form. The former two vector fields are used to realize decoupled motion control of the end-effector and the arm plane, respectively. No additional singularities are introduced, and motion through kinematic singularities is stable and cyclic.

The shortage of human resources on the International Space Station (ISS) is becoming a serious problem. To tackle this problem, smart robot systems that support human activities in space should be developed. In this paper, a new space robot system named "Intra-Vehicular Free-Flyer System (IVFFS)" is proposed. The IVFFS supports not only non-contact but also contact tasks during intra-vehicular activities (IVA). To accomplish such requirements, we introduce a prototype model called "Space Humming Bird" (SHB) which has a variably structured body to satisfy both safety and dexterity requirements. Furthermore, several SHBs can be combined to achieve more complicated tasks. To confirm the feasibility of our concept, a CG simulator is developed. In addition, several operator support techniques are incorporated into the tele-operation system.

The SC method is applied to the positioning sub-chain of a seven-DOF with zero joint offsets, such as the MHI PA-10 arm. Via the two vector fields obtained, decoupled motion control of the end-effector and the arm plane, respectively, can be realized. No additional singularities are introduced, and motion through kinematic singularities is stable.

We compare the main features of the singularity-consistent (SC) method and the damped-least-squares (DLS) method. It is shown that both methods introduce a so-called algorithmic error in the vicinity of a singular point. The direction of this error is, however, different in each method. This is shown to play an important role for system stability.

This article presents impact dynamic analysis of a free-floating space robot, subject to a force impulse at the hand. We study the joint and the base reactions in terms of finite velocity changes and clarify their role for the post-impact motion behavior of the robot. The analysis makes use of a joint-space orthogonal decomposition procedure involving the so called reaction null space. The article focuses on the specific case of a nonredundant arm and a reaction null space in terms of base angular motion. We further show that with proper post-impact control it is possible to transfer the whole angular momentum from the base toward the manipulator, and in the same time to reduce the joint velocity.

A composite control law for end-effector path tracking with a flexible structure mounted manipulator system is proposed, such that no disturbances on the flexible base are induced. The control law is based on the reaction null-space concept introduced earlier to tackle dynamic interaction problems of free-floating robots, or moving base robots in general. The control law is called composite since it ensures base vibration suppression control as well, although independently of the reactionless motion control subtask. The requirement of task independence is essential to avoid the appearance of complex dynamics expressions in the control law, such as nonlinear velocity-dependent coupling terms and dependencies of inertias on the elastic coordinates. We present experimental data from computer simulations and the experimental test bed TREP developed at Tohoku university. The experimental data is shown to agree well with theory.

This work is an extension of the authors' previous result (1995), mainly to tackle the post-impact control problem. We focus on the specific case of a nonredundant arm and a reaction null space in terms of base angular motion. It is shown that with proper post-impact manipulator control it is possible to swiftly transfer the whole angular momentum from the base toward the manipulator, and in the same time to reduce the joint velocity.

We discuss a new type of space robot system composed of a spacecraft and a robot attached through a tether to it. The tethered robot is translated away from the spacecraft using existing tether tension control techniques in gravitational field. Link motion of the tethered robot, however, is complicated, since the momentum is not constant due to the presence of external forces. This paper especially focuses on momentum control of the tethered robot. In addition to translational momentum control, we show that angular momentum of the tethered robot can be controlled by proper motion of the tether attachment point. We propose a control law for link motion of the tethered robot composed of two subtasks: the end-effector motion subtask and the tether attachment point motion subtask.

In this paper we propose a new compact 6-DOF haptic interface with large workspace. It contains a newly developed five bar spatial gimbal mechanism for orientation, placed on a modified Delta parallel-link mechanism. The motion range of each axis of the five bar mechanism is over /spl plusmn/70 degrees. Quick motions can be realized easily due to the parallelism inherent to both the modified Delta substructure and the five bar substructure.

This work reports progress on a long-reach manipulator project. The original single-arm manipulator was complemented with an identical second arm. We introduce several noncontact motion control strategies which are based on the reaction null space concept. Experimental verification of disturbance compensation control via a single arm, and via the two arms while holding an object, is done. Also, motion feasibility on reactionless paths for a closed kinematic chain, including the two arms and the object, is examined.

In our previous work (1997, 1998) we proposed two approaches to singularity treatment: one is based on a null space notation, and the other on the adjoint Jacobian. We use the term "singularity-consistent (SC) approach" in referring to any of them. The SC approach guarantees the direction of motion in the whole work space. Thus, it becomes possible to move through singularities in teleoperated mode, without producing an infeasible joint velocity. However, a precise command direction is needed to accomplish such a specific motion. In this paper, we propose a supporting system for the operator to handle the above problem. An advanced SC teleoperation system is obtained, based on virtual reality techniques. Experimental results show that a real peg-in-hole task, including through-singularity motion, can be easily realized by the operator.

A general framework for under-actuated manipulator systems is introduced. Within this framework, we show how to decompose the second-order dynamic motion constraint into two orthogonal components. Based on this decomposition, feedback control laws are proposed for motion stabilization to a reactionless-motion equilibrium manifold. Reactionless motion without drift is guaranteed for first-order nonholonomic systems. It is also shown that for a second-order nonholonomic system, reactionless motion in general leads to a drift.

A singularity-consistent direct kinematics algorithm is a necessity for the analysis and control of parallel manipulators. The present work proposes a first order singularity consistent algorithm for a class of parallel manipulators. It is shown that the algorithm is stable, convergent and can handle the multiplicity of the direct kinematics solutions for the manipulators. The performance of the method is analyzed with the help of the numerical examples given.

This work adds some new insight into the singularity-consistent parametrization of the inverse kinematics of a nonredundant robot tracing a reference path. We show that in the general case (e.g. motion with constant end-effector velocity on the path) there is a nonintegrable motion component. When the end-effector moves with a velocity proportional to the determinant of the Jacobian, the nonintegrable motion component is removed. At the same time, the order of the equation of motion is reduced. We call this type of motion "natural". Furthermore, from the simplified equation of natural motion, we derive the mechanical power measure. This is an instantaneous-motion performance measure defined over the phase space of the system, and thus, can be useful to evaluate the dynamic performance for the specified instantaneous motion direction.

The control of a dextrous manipulator mounted on a flexible structure is discussed. Using the concept of reaction null space, the manipulator dynamics is decoupled from the base dynamics. As a consequence of the decoupling, feedback control gains for structural vibration suppression and manipulator end-point control can be determined in a straightforward manner. We examine experimentally the performance of the above control tasks, using a planar experimental setup.

In our previous study we introduced a new type of space robot system, which consists of a spacecraft and a robot attached to it via a tether. In this paper, the characteristics of the system in gravitational field are described and a path planning approach is proposed. The translational motion of the mass center of the robot is controlled by tether tension. The path planning approach ensures end-effector motion along the desired path. In addition, proper motion of the tether attachment point is obtained, such that tension control can take effect. The effectiveness of the approach is examined by computer simulations.

The singularity-consistent path-planning and motion control approach has so far been successfully tested within a teleoperation environment, but with separate control of the positioning subsystem and the wrist of the slave arm. In this paper, we show that the approach can be applied to a 6-DOF slave arm with a non-spherical wrist. The problems arising from the coupling between position and orientation are handled easily within the null-space/adjoint Jacobian framework of the singularity-consistent approach. The experiments show that the operator is able to approach the shoulder singularity smoothly, to move through it or to perform motions without leaving it.

We analyze some aspects of human physical skill and propose a model thereof, with a view to implementing such skills in robots. We implement our model in a control scheme which makes use of impedance control in combination with low-gain force control and feedforward control. The summation results show that the model is able to achieve the reference force, and at the same time, the manipulator shows impedance-type behavior. Furthermore, with this scheme it is possible to track the reference force quickly, when the disturbance is known apriori. The experiments show that the performance of our control scheme is very close to that of the human operator.

Proposes a new technique for closed-loop kinematic control of nonredundant robotic mechanisms, based on the adjoint matrix of the kinematic Jacobian. Using the Lyapunov direct method, the authors show that the adjoint Jacobian approach guarantees asymptotic stability at regular points, around singularities, and at so-called instantaneous self-motion singularities. The new property, as compared to previous approaches, is that direction of motion can be precisely controlled at those points. To guarantee the asymptotic stability around any singularity and at instantaneous self-motion singularities, the desired (scalar) end-effector velocity is appropriately modified, and at the same time, restriction on the joint velocity norm according to a user-specified valve is imposed. In the vicinity of a singularity an error in the position along the desired path is tolerated, which however, does not lead to deviation from the path.

In this paper we discuss the relation between the two approaches to velocity command generation for nonredundant robotic mechanisms, which the two groups of the authors proposed recently and independently of each other. It will be shown analytically that the singularity-consistent null space based approach, and the split Jacobian approach, are equivalent. Analysis of the behavior at a singularity will be presented from the viewpoint of both approaches. An analytical example will be used to demonstrate the theoretical results.

We apply our newly proposed singularity-consistent path tracking approach to nonredundant parallel-link manipulators. We analyze the singularities of such mechanisms, assuming that the output-link moves on a pre-defined and parametric path. Especially, we focus on the so-called instantaneous self-motion type singularity. We propose a closed-loop controller that guarantees asymptotic stability when tracking paths through such a singularity. As a comprehensive analytical example we use a planar five bar mechanism. A computer simulation study is also presented, using the same example, as well as the HEXA parallel robot structure.

We discuss the implementation and experimental verification of two new approaches to control of a nonredundant manipulator at and around singularities. One of the approaches is based on the adjoined matrix of the manipulator Jacobian, while the other one is a null-space based approach. The approaches share a common theoretical base described in our previous work. We focus mainly on several implementation issues and the experimental results obtained from our teleoperation system.

In this paper, we deal with the handling of a flexible object by rigid arm robots. We consider three main tasks: 1) to propose a mathematical model for a variety of flexible objects of our daily life; 2) to design a controller to achieve cooperative handling of the flexible object by the robots; and 3) to analyze the stability and robustness of the control system. In particular, demands for manipulating a large-scale structure by a space robot will be increasing. Therefore, it is important to constitute the cooperative control problem of several robots handling a flexible object, and to analyze the proposed control system.

We develop further our singularity-consistent approach to arrive at a parameterized form of the dynamics of a nonredundant robotic mechanism tracking a desired path in Cartesian space. It is shown that this form is suitable for incorporating joint torque limits, which is an important issue for practical applications. We propose a closed-loop controller which behaves as a "conventional" resolved-acceleration type controller at regular points of the kinematic function. Around any singularity and at so-called instantaneous self-motion singularities the controller is able to truck the direction of the specified path exactly. The limit on the torque norm results in some position error, without deteriorating, however, the direction tracking ability. It is shown also that motion through the bifurcation type singularity can be easily controlled in practice as well.

For a class of parallel-link manipulators we develop a general formulation of the equation of motion, suitable for parallel computations. We analyze the torque requirement when moving through various self-motion type singularities on a path generated under the singularity-consistent framework. The formulation, contributes mainly to the analysis of a singularity which is typical for parallel robots only, known from previous studies as "overmobility". We show that if the dynamics of the system is taken, under consideration, it is possible to move through such a singularity. This analysis motivates the introduction of the concept of dynamic singularity consistency. As a comprehensive analytical example we use a five bar robotic mechanism.

We propose a new singularity-perturbed design approach to a planar five bar parallel-link manipulator. The design allows us to operate the manipulator either in parallel or in serial branch mode, and also to exchange those modes. Thus, it is possible to merge some of the well-known advantages of serial and parallel manipulators. We study two basic singularity-perturbed designs and show that one of them is preferable. A feasibility study through computer simulation, including maximum torque requirement, is also presented.

The authors previously (1996) proposed a new type of space robot system, consisting of a spacecraft and a robot attached through a tether to it. This system is called a "tethered robot system." In this paper, we clarify the characteristics of the tethered robot system, and especially focus on position control of the center of mass of the robot. Tether tension is used to control the position, taking into account the gravity gradient and the centrifugal force. The results from the trajectory planning procedure suggest that the shape of the path depends both on the direction to the destination point and the time to accomplish the mission, it does not depend on the distance. From the simulation results with feedforward control it is concluded that accurate path planning is possible only if the destination point is close to the initial point, or if the time to accomplish the mission is long enough.

Point-to-point operation of a flexible structure mounted manipulator systems (FSMS) is discussed. Four operation strategies: (1) straight-line path in joint space, (2) high-coupling path, (3) low-coupling path obtained from the coupling map concept, and (4) 3-phase motion obtained from the reactionless path are examined and compared in terms of a minimum oscillation of the supporting flexible structure, using an FSMS test bed, TREP, developed at Tohoku University.

In this paper we develop further the recently proposed null-space method for path tracking at and around kinematic singularities. We consider two types of singularities known as ordinary singularities and bifurcation/isolated-point singularities. A closed-loop kinematic control scheme is introduced, which is able to keep precisely the direction of the specified end-effector path passing arbitrary close to kinematic singularities. The desired velocity magnitude can be maintained at regular points of the kinematic mapping. In the vicinity of singularities, the maximum available joint velocity specified from hardware limits can be applied. Several advantages of the proposed singularity-consistent approach are pointed out when compared with other well-known methods.

A general method for local kinematic inversion of non-redundant and kinematically redundant robotic mechanisms is proposed. The mathematical background is a recursive scheme based on gradient projection through pseudoinverses. The additional task (in case of a kinematically redundant robotic mechanism) and/or the end-effector tasks are decomposed into single task components. Priority among these components is allocated dynamically rather than keeping it fixed, as in other schemes. This approach yields the advantage of "sacrificing" only the worst-conditioned task components. Further on, it is shown how the notation can be modified to damp single solution components in the neighbourhood of task and/or algorithmic singularities, in order to guarantee some bounded solution norm.>

This paper presents a flexible redundancy resolution strategy based on the task-priority method. A dynamic task-priority allocation approach has been motivated by the fact that the performance may degenerate for improper fixed-priority assignment to various task components. Recursive local kinematic inversion is applied, which, along with a full task-decomposition approach, guarantees the flexibility of the approach. It is further shown that the damping technique is easily implemented yielding a scheme that performs well also in the neighborhood of singularities. Thereby, the computationally inefficient singular-value-decomposition has been avoided.>

A resolved acceleration type controller for free-flying space robots with a kinematically redundant manipulator arm is presented. The arm is controlled to track a desired end-effector trajectory and at the same time to change the attitude of the system in a desired manner and without activating jet thrusters and/or reaction wheels. The formulation is based on the fixed-attitude restricted (FAR) Jacobian matrix which was introduced for path planning and control at the velocity level. A reformulation in terms of accelerations allows the author to address the issue of end-effector velocity step-change response, caused, for example, by a collision between the end-effector and the target. It is shown that when the FAR Jacobian is applied, the joint acceleration is mainly derived from the null space of the manipulator-link inertia matrix, and hence there is very little disturbance of the spacecraft attitude. As a consequence, there is a loose dependence on spacecraft mass and inertia.

An analysis of the momentum conservation equations of a redundant free-flying spacecraft/manipulator system acting in a zero-gravity environment is presented. In order to follow a predefined end-effector path, the inverse kinematics at velocity level is considered. The redundancy is solved alternatively in terms of pseudoinverses and null-space components of the manipulator inertia matrix, the manipulator Jacobian matrix, and the generalized Jacobian matrix. A general manipulation task is defined as end-effector continuous path tracking with simultaneous attitude control of the spacecraft. Three subtasks of the general task are considered. The case of manipulator motions that yield no spacecraft attitude disturbance is analyzed in more detail and a special 'fixed-attitude-restricted' (FAR) Jacobian is defined. Through singular-value decomposition of this Jacobian, corresponding FAR dexterity measures (FAR manipulability and FAR condition number) are derived.>