Showing papers in "The International Journal of Robotics Research in 1999"
TL;DR: This paper algebraically prove that if the authors consider the camera and motor transformations as screws, then only the line coefficients of the screw axes are relevant regarding the hand-eye calibration, and shows how a line transformation can be written with the dual-quaternion product.
Abstract: To relate measurements made by a sensor mounted on a mechanical link to the robot’s coordinate frame, we must first estimate the transformation between these two frames. Many algorithms have been p...
681 citations
TL;DR: Several strategies are presented for stabilizing the Passive dynamic walking motion, of which the quasi-static control of step width is determined to be both simple and efficient.
Abstract: Passive dynamic walking refers to a class of bipedal machines that are able to walk down a gentle slope with no external control or energy input. The legs swing naturally as pendula, and conservation of angular momentum governs the contact of the swing foot with the ground. Previous machines have been limited to planar motions. We extend the planar motions to allow for tilting side to side (roll motion). Passive walking cycles exist, but the roll motion is unstable, resembling that of an inverted pendulum. The instability is due to mismatching of roll velocity with the ground contact conditions. Several strategies are presented for stabilizing this motion, of which the quasi-static control of step width is determined to be both simple and efficient.
624 citations
TL;DR: The goal is to develop a manipulation system with the precision and sensitivity of a machine, but with the manipulative transparency and immediacy of hand-held tools for tasks characterized by compliant or semi-rigid contacts with the environment.
Abstract: This paper reports the development of a robotic system designed to extend a human’s ability to perform small-scale (sub-millimeter) manipulation tasks requiring human judgment, sensory integration,...
568 citations
TL;DR: In this article, the problem of foot rotation in biped robots during the single-support phase was studied and it was shown that foot rotation is an indication of postural instability.
Abstract: The focus of this paper is the problem of foot rotation in biped robots during the single-support phase. Foot rotation is an indication of postural instability, which should be carefully treated in...
559 citations
TL;DR: An implementation of stochastic mapping that uses a delayed nearest neighbor data association strategy to initialize new features into the map, match measurements to map features, and delete out-of-date features is described.
Abstract: The task of building a map of an unknown environment and concurrently using that map to navigate is a central problem in mobile robotics research. This paper addresses the problem of how to perform concurrent mapping and localization (CML) adaptively using sonar. Stochastic mapping is a feature-based approach to CML that generalizes the extended Kalman filter to incorporate vehicle localization and environmental mapping. The authors describe an implementation of stochastic mapping that uses a delayed nearest neighbor data association strategy to initialize new features into the map, match measurements to map features, and delete out-of-date features. The authors introduce a metric for adaptive sensing that is defined in terms of Fisher information and represents the sum of the areas of the error ellipses of the vehicle and feature estimates in the map. Predicted sensor readings and expected dead-reckoning errors are used to estimate the metric for each potential action of the robot, and the action that yi...
373 citations
TL;DR: Nonrobotics applications (e.g., graphic animation, surgical planning, computational biology) are growing in importance and are likely to shape future motion-planning research more than robotics itself.
Abstract: During the past three decades, motion planning has emerged as a crucial and productive research area in robotics. In the mid-1980s, the most advanced planners were barely able to compute collision-free paths for objects crawling in planar workspaces. Today, planners efficiently deal with robots with many degrees of freedom in complex environments. Techniques also exist to generate quasioptimal trajectories, coordinate multiple robots, deal with dynamic and kinematic constraints, and handle dynamic environments. This paper describes some of these achievements, presents new problems that have recently emerged, discusses applications likely to motivate future research, and finally gives expectations for the coming years. It stresses the fact that nonrobotics applications (e.g., graphic animation, surgical planning, computational biology) are growing in importance and are likely to shape future motion-planning research more than robotics itself.
275 citations
TL;DR: A one-joint manipulator is developed that can perform a variety of dynamic tasks, including snatching an object ftom a table, rolling an object on the surface of the arm, and throwing and catching, and nonlinear optimization is used to plan robot trajectories that achieve the desired object motion via coupling forces though the nonprehensile contact.
Abstract: We are interested in using low-degree-of-freedom robots to perform complex tasks by nonprehensile manipulation (manipulation without aformorforce-closure grasp). By notgrasping, the robot can usegravitational, centrifugal, and Coriolisforces as virtual motors to control more degrees of freedom of the part. The part s extra motionfreedoms are exhibited as rolling, slipping, and free flight.This paper describes controllability, motion planning, and implementation ofplanar dynamic nonprehensile manipukltion. We show that almost any planar object is controllable by point contact, and the controlling robot requires only twvo degrees of freedom (a point translating in the plane). We then focus on a one-joint manipulator (with a two-dimensional state space), and show that even this simplest of robots, by using slipping and rolling, can control a planar object to a fulldimensional subset of its six-dimensional statespace. We have developed a one-jointrobotto perform a variety of dynamic tasks, including snatching...
271 citations
TL;DR: It is shown that framewalkers are appropriate for rappelling in severe terrain, though tether systems have limitations, and the importance of future “autonomous” systems to realize when they require human support rather than relying on humans for constant oversight is discussed.
Abstract: Dante II is a unique walking robot that provides important insight into high-mobility robotic locomotion and remote robotic exploration. Dante II’s uniqueness stems from its combined legged and rappelling mobility system, its scanning-laser rangefinder, and its multilevel control scheme. In 1994 Dante II was deployed and successfully tested in a remote Alaskan volcano, as a demonstration of the fieldworthiness of these technologies. For more than five days the robot explored alone in the volcano crater using a combination of supervised autonomous control and teleoperated control. Human operators were located 120 km distant during the mission. This article first describes in detail the robot, support systems, control techniques, and user interfaces. We then describe results from the battery of field tests leading up to and including the volcanic mission. Finally, we put forth important lessons which comprise the legacy of this project. We show that framewalkers are appropriate for rappelling in severe terr...
264 citations
TL;DR: A novel feedback-control law for coordinating the motion of multiple holonomic mobile robots to capture/enclose a target by making troop formations is presented, which is a cooperative behavior for security against invaders in surveillance areas.
Abstract: This paper presents a novel feedback-control law for coordinating the motion of multiple holonomic mobile robots to capture/enclose a target by making troop formations. This motion coordination is a cooperative behavior for security against invaders in surveillance areas. Each robot in this control law has its own coordinate system and it senses a target/invader, other robots and obstacles, to achieve this cooperative behavior without making any collision. Although there is no centralized controller and each robot has local feedback that is relative-position feedback, all the robots are asymptotically stabilized, and they make formations enclosing a target. Each robot especially has a vector referred to as a “formation vector,” and the formations are controllable by the vectors. As for determining the formation vectors, we use a reactive-control framework in which robots have some reactions heuristically designed according to this cooperative behavior. Therefore, this robotic system is a hybrid system tha...
259 citations
TL;DR: A new theory in contact mechanics for modeling of soft fingers is proposed to define the relationship between the normal force and the radius of contact for soft fingers by considering general soft-finger materials, including linearly and nonlinearly elastic materials.
Abstract: A new theory in contact mechanics for modeling of soft fingers is proposed to define the relationship between the normal force and the radius of contact for soft fingers by considering general soft-finger materials, including linearly and nonlinearly elastic materials. The results show that the radius of contact is proportional to the normal force raised to the power of, which ranges from 0 to 1/3. This new theory subsumes the Hertzian contact model for linear elastic materials, where D 1/3. Experiments are conducted to validate the theory using artificial soft fingers made of various materials such as rubber and silicone. Results for human fingers are also compared. This theory provides a basis for numerically constructing friction limit surfaces. The numerical friction limit surface can be approximated by an ellipse, with the major and minor axes as the maximum friction force and the maximum moment with respect to the normal axis of contact, respectively. Combining the results of the contactmechanics mo...
238 citations
TL;DR: The article presents developments of models, strategies, and algorithms concerned with a number of autonomous capabilities that are essential for robot operations in human environments, including integrated mobility and manipulation, cooperative skills between multiple robots, interaction ability with humans, and efficient techniques for real-time modification of collision-free path.
Abstract: This article discusses the basic capabilities needed to enable robots to operate in human-populated environments for accomplishing both autonomous tasks and human-guided tasks. These capabilities are key to many new emerging robotic applications in service, construction, field, underwater, and space. An important characteristic of these robots is the “assistance” ability they can bring to humans in performing various physical tasks. To interact with humans and operate in their environments, these robots must be provided with the functionality of mobility and manipulation. The article presents developments of models, strategies, and algorithms concerned with a number of autonomous capabilities that are essential for robot operations in human environments. These capabilities include: integrated mobility and manipulation, cooperative skills between multiple robots, interaction ability with humans, and efficient techniques for real-time modification of collision-free path. These capabilities are demonstrated ...
TL;DR: An overview of the new recursive, divide-and-conquer algorithm for calculating the forward dynamics of a robot mechanism, or general rigid-body system, is presented and a detailed description of the simplest case: unbranched kinematic chains is presented.
Abstract: This paper presents a recursive, divide-and-conquer algorithm for calculating the forward dynamics of a robot mechanism, or general rigid-body system, on a parallel computer. It features O(log(n)) time complexity on O(n) processors and is the fastest available algorithm for a computer with a large number of processors and low communications costs. It is an exact, noniterative algorithm and is applicable to mechanisms with any joint type and any topology, including branches and kinematic loops. The algorithm works by recursive application of a formula that constructs the articulatedbody equations of motion of an assembly from those of its constituent parts. The inputs to this formula are the equations of motion of two independent subassemblies, plus a description of how they are to be connected, and the output is the equation of motion of the assembly. Starting with a collection of unconnected rigid bodies, the equations of motion of any rigid-body system can be constructed by repeated application of this ...
TL;DR: An analysis of the problem of steering an autonomous vehicle along a highway based on the images obtained from a CCD camera mounted in the vehicle by exploring the effects of changing various important system parameters like the vehicle velocity, the look-ahead range of the vision sensor, and the processing delay associated with the perception and control systems.
Abstract: With the increasing speeds of modern microprocessors, it has become ever more common for computer-vision algorithms to find application in real-time control tasks. In this paper, we present an analysis of the problem of steering an autonomous vehicle along a highway based on the images obtained from a CCD camera mounted in the vehicle. We explore the effects of changing various important system parameters like the vehicle velocity, the look-ahead range of the vision sensor, and the processing delay associated with the perception and control systems.We also present the results of a series of experiments that were designed to provide a systematic comparison of a number of control strategies. The control strategies that were explored include a lead-lag control law, a full-state linear controller, and an input-output linearizing control law. Each of these control strategies was implemented and tested at highway speeds on our experimental vehicle platform, a Honda Accord LX sedan.
TL;DR: Algorithms are presented to determine all the possible locations of the center of the platform that can be reached with a fixed orientation and the inclusive orientation workspace, which shows that for robots of similar dimensions the joints layout has a large influence on the workspace volume.
Abstract: We consider in this paper a Gough-type parallel robot whose leg length values are constrained to lie within some fixed ranges and for which there may be mechanical limits for the motion of the passive joints.The purpose of this paper is to present algorithms to determine:• the constant orientation workspace: all the possible locations of the center of the platform that can be reached with a fixed orientation• thetotal orientation workspace: all the possible locations of the center of the platform that can be reached with any orientation in a set defined by three ranges for the orientation angles (the dextrous workspace is an example of total orientation workspace case, the three ranges being T [0,360] degree1)• the inclusive orientationworkspace: all the possible locations of the center of the platform that can be reached with at least one orientation among a set defined by three ranges for the orientation angles (the maximal or reachableworkspace is an example of inclusive orientation workspace, the thre...
TL;DR: A more accurate version of the algorithm is presented and the results of some numerical accuracy tests that compare both versions with the standard articulated body algorithm are presented.
Abstract: This paper is the second in a two part series describing a recursive, divide and conquer algorithm for calculating the forward dynamics of a robot mechanism, or a general rigid body system, on a pa...
TL;DR: This paper presents a strategy for achieving practical mapping navigation using a wheeled mobile robot equipped with an advanced sonar sensor that drastically improves the memory and processing-time requirements of the original global approach, while preserving the statistical details necessary for an accurate map and prolonged navigation.
Abstract: This paper presents a strategy for achieving practical mapping navigation using a wheeled mobile robot equipped with an advanced sonar sensor The original mapping navigation experiment, carried out with the same robot configuration, built a feature map consisting of common-place indoor landmarks crucial for localization; namely, planes, corners, and edges. The map exhaustively maintained covariance matrices among all features, and thus presented a time and memory impediment to practical navigation in large environments. The new local mapping strategy proposed here breaks down a large environment into a topology of local regions, only maintaining the covariance among features in the same local region, and the covariance among local maps. This notion of two-hierarchy representation drastically improves the memory and processing-time requirements of the original global approach, while preserving the statistical details, in the authors' opinions, necessary for an accurate map and prolonged navigation. The new...
TL;DR: Using stratified Morse theory, it is shown that the hand’s configuration at which the cage is broken corresponds to a frictionless equilibrium grasp, which allows for a technique for computing the caging set of a two-fingered hand whose opening is controlled by a single parameter.
Abstract: This paper is concerned with a caging problem: we wish to surround an object B by a multifingered hand such that B has some freedom to move, but still cannot escape the “cage” formed by the fingers. We introduce a new notion of the caging set, which is based on the configuration-space representation of the free motions of the hand system with respect to the object. Using stratified Morse theory, we show that the hand’s configuration at which the cage is broken corresponds to a frictionless equilibrium grasp. This allows us to formulate a technique for computing the caging set of a two-fingered hand whose opening is controlled by a single parameter. The tech-nique generalizes to one-parameter grip ping systems having a higher number of fingers.
TL;DR: The solution of minimum-effort optimal control problems for open-chain manipulators is examined and it is demonstrated that the parameter-optimization formulation of the problem is numerically ill-conditioned, and that it is therefore essential to include analytic gradients of the objective fiunction and the constraints in order to guarantee a solution.
Abstract: In this article, we examine the solution of minimum-effort optimal control problems for open-chain manipulators. An approximate Solution to the optimal control problem is determined by a constrained parameter optimization over a set of B-spline basis functions. We demonstrate that the parameter-optimization formulation of the problem is numerically ill-conditioned, and that it is therefore essential to include analytic, or exact, gradients of the objective fiunction and the constraints in order to guarantee a solution. A recursive expression for these gradients is developed for general serial chains. Constraints on end-effector motions are taken into account using the logarithmn of the spatial displacement. Our formulation relies on the use of matrix exponentials for the manipulator kinematics, dynamics, and task constraints. Several examples are presented that demostrate the power and flexibility of our approach.
TL;DR: This paper uses (linearized) Kalman filters to estimate first-order geometric parameters that occur in force-controlled compliant motions, and uses the reciprocity constraint between ideal contact forces and motion freedoms as the “measurement equation” of the Kalman filter.
Abstract: This paper uses (linearized) Kalman filters to estimate first-order geometric parameters (i.e., orientation of contact normals and location of contact points) that occur in force-controlled compliant motions. The time variance of these parameters is also estimated. In addition, transitions between contact situations can be monitored. The contact between the manipulated object and its environment is general, i.e., multiple contacts can occur at the same time, and both the topology and the geometry of each single contact are arbitrary. The two major theoretical contributions are (1) the integration of the general contact model, developed previously by the authors, into a state-space form suitable for recursive processing; and (2) the use of the reciprocity constraint between ideal contact forces and motion freedoms as the measurement equation of the Kalman filter. The theory is illustrated by full 3-D experiments. The approach of this paper allows a breakthrough in the state of the art dominated by the classical, orthogonal contact models of Mason that can only cope with a limited (albeit important) subset of all possible contact situations.
TL;DR: Results from studying the kinematic geometry of such systems are reviewed, and the three most useful solution techniques are summarized.
Abstract: The kinematic analyses, of manipulators and other robotic devices composed of mechanical links, usually depend on the solution of sets of nonlinear equations. There are a variety of both numerical and algebraic techniques available to solve such systems of equations and to give bounds on the number of solutions. These solution methods have also led to an understanding of how special choices of the various structural parameters of a mechanism influence the number of solutions inherent to the kinematic geometry of a given structure. In this paper, results from studying the kinematic geometry of such systems are reviewed, and the three most useful solution techniques are summarized. The solution techniques are polynomial continuation, Grobner bases, and elimination. We then discuss the results that have been obtained with these techniques in the solution of two basic problems, namely, the inverse kinematics for serial-chain manipulators, and the direct kinematics of in-parallel platform devices.
TL;DR: It is shown that never-special assembly changing motions can be excluded by making platform and base triangles similar, and it is conjecture that appropriate, perhaps identical, specialization for the octahedral manipulator has the same effect.
Abstract: When the leg rods of a fully in-parallel manipulator are fixed in their lengths, it is usual that the device can be assembled in several distinct ways. Sometimes it happens that motion between such assemblies can take place such that the linkage is never at a special configuration; that is, a configuration where the moving-platform body acquires uncontrollable freedom relative to the base. The possibility of such motion has implications for control. Focusing on 3-3 devices, the authors present a geometric explanation of how these motions arise, and give a sufficient condition for their existence. For the 3-3 planar-motion device, they show that never-special assembly changing motions can be excluded by making platform and base triangles similar, and they conjecture that appropriate, perhaps identical, specialization for the octahedral manipulator has the same effect.
TL;DR: This paper provides a theoretical framework for controlling a manipulator with hyper degrees of freedom (HDOF) and gives theorems on the existence of the solutions, and on an existence region that allows us to convert shape-control problems into more tractable ones.
Abstract: This paper provides a theoretical framework for controlling a manipulator with hyper degrees of freedom (HDOF) . An HDOF manipulator has the capability to achieve various kinds of tasks. To make full use of its capability, shape control is proposed here; that is, not only the tip of a manipulator, but also its whole body is controlled. To formulate control objectives for shape control, the authors define a shape correspondence between an HDOF manipulator and a spatial curve that prescribes a desired shape. The shape correspondence is defined by using solutions of a nonlinear optimization problem termed the shape-inverse problem. They give theorems on the existence of the solutions, and on an existence region that allows them to convert shape-control problems into more tractable ones. A shape-regulation control problem is considered first to bring an HDOF manipulator onto a given time-invariant curve. The idea of estimating the desired curve parameters is the crucial key to solving the problem by Lyapunov design. The derived shape-regulation law includes the estimator, which infers the desired curve parameters corresponding to the desired joint positions on the curve. The idea of the desired curve-parameter estimation is also effective for shape tracking where a time-varying curvemore » is used for prescribing a moving desired shape. Considering an estimator with second-order dynamics enables the authors to find two shape-tracking control laws by utilizing conventional tracking methods in manipulator control. They show the simulation results of applying the derived shape-tracking control laws to a 20-DOF manipulator.« less
TL;DR: The hardware design, control, and navigation system of and some preliminary experiments with the robotic wheelchair Mobility Aid for elderly and disabled people, which has survived more than 36 h of testing in public, crowded environments with heavy passenger traffic.
Abstract: In this paper we describe the hardware design, the control and navigation system, and our preliminary experiments with the robotic wheelchair MAid (Mobility Aid for Elderly and Disabled People).
TL;DR: This paper surveys work on a single joint robot that orients parts on a conveyor belt; a robot that uses dynamics to snatch, roll, or throw objects; hitting things to position them; manipulating things whose shapes are not completely known; and integration of manipulation with locomotion.
Abstract: This paper reviews my recent research in robotic manipulation and speculates on potentially fruitful directions for future work. My recent work is focused on nonprehensile manipulation: manipulating objects without grasping them. In particular, the paper surveys work on a single joint robot that orients parts on a conveyor belt; a robot that uses dynamics to snatch, roll, or throw objects; hitting things to position them; manipulating things whose shapes are not completely known; and integration of manipulation with locomotion. In the future, a broad view of robotics will allow us to focus on fundamental principles and at the same time address a variety of new applications.
TL;DR: This paper describes new experimental devices that can implement programmable force fields, and describes new manipulation algorithms, including the M-CHIP (Manipulation CHIP), a massively parallel array of programmable micromotion pixels.
Abstract: Programmable force vector fields can be used to control a variety of flexible planar parts feeders such as massively parallel microactuator arrays or transversely vibrating (macroscopic) plates. Th...
TL;DR: This work designs a system of two agents to perform a cooperative navigation task and shows how competition among constraints enables agents to make decisions regarding which behavior to execute in a given situation, resulting in the execution of sequences of behaviors that satisfy task requirements.
Abstract: The dynamic-systems approach to robot path planning defines a dynamics of robot behavior in which task constraints contribute independently to a nonlinear vector field that governs robot actions. The authors address problems that arise in scaling this approach to handle complex behavioral requirements. They propose a dynamics that operates in the space of task constraints, determining the relative contribution of each constraint to the behavioral dynamics. Competition among task constraints is able to deal with problems that arise when combining constraint contributions, making it possible to specify tasks that are more complex than simple navigation. To demonstrate the utility of this approach, the authors design a system of two agents to perform a cooperative navigation task. They show how competition among constraints enables agents to make decisions regarding which behavior to execute in a given situation, resulting in the execution of sequences of behaviors that satisfy task requirements. The authors discuss the scalability of the competitive-dynamics approach to the design of more complex autonomous systems.
TL;DR: A mobile robot equipped with a sonar sensor array in a guided, feature-based map-building task in an indoor environment and a comparison is made with the Julier-Uhlmann-Durrant-Whyte Kalman filter which improves the accuracy of covariance propagation when nonlinear equations are involved.
Abstract: This paper describes a mobile robot equipped with a sonar sensor array in a guided, feature-based map-building task in an indoor environment. The landmarks common to indoor environments areplanes, corners, and edges, and these are located and classified with the sonar sensor array The map-building process makes use of accurate odometry information that is derivedfiom apair ofknife-edged unloaded encoder wheels. Discrete sonar observations are incrementally merged into partial planes to produce a realistic representation of the environment that is amenable to sonar localization. Collinearity constraints among featurs ar exploited to enhance both the map-featwe estimation and robot localization. The map update employs an iterated extended Kalmanfilter in the first implementation and subsequently a comparison is made with the Julier-Uhlmann-Durrant-Whyte Kalman filter which improves the accuracy of covariance propagation when nonlinear equations are involved. The map accounts for correlation among features a...
TL;DR: The differential control for finger tracking is described and analyzed and extended to on-line continuous control for a set of cooperating robot fingers and is computationally efficient, exact, and takes into consideration the full dynamics of the system.
Abstract: We present an algorithm called finger tracking for in-hand manipulation of three-dimensional objects with independent robot fingers. We describe and analyze the differential control for finger trac...
TL;DR: A path-tracking criterion for the so-called LHD (load-haul-dump) truck used in underground mining is proposed in this paper, based on monitoring the distances of the midpoints of both axles of the vehicle from their orthogonal projections on the path, using two different moving frames simultaneously.
Abstract: A path-tracking criterion for the so-called LHD (load-haul-dump) truck used in underground mining is proposed in this paper. It exploits the particular configuration of this vehicle, composed of two units connected by an actuated articulation. The task is to follow the path represented by the middle of the tunnel, maintaining the whole vehicle at a reduced distance from the path itself, to decrease the risk of crashes against the walls of the tunnel. This is accomplished via feedback through the synthesis of an appropriate path-tracking criterion. The criterion is based on monitoring the distances of the midpoints of both axles of the vehicle from their orthogonal projections on the path, using two different moving frames simultaneously. Local asymptotic stability to paths of constant curvature is achieved by means of linear-state feedback.
TL;DR: A key lesson is that system robustness can only be ensured by exhaustively exercising the system’s operational capabilities in a number of diverse environments.
Abstract: The Mobile Detection Assessment and Response System robotic security program has successfully demonstrated simultaneous control of multiple robots navigating autonomously within an operational warehouse environment. This real-world warehouse installation required adapting a navigational paradigm designed for highly structured environments such as office corridors (with smooth walls and regularly spaced doorways) to a semistructured warehouse environment (with no exposed walls and within which odd-shaped objects unpredictably move about from day to day). A number of challenges, some expected and others unexpected, were encountered during the transfer of the system first to a beta-test/demonstration site and then to an operational warehouse. This paper examines these problems (and others previously encountered) in a historical context of the evolution of navigation and other needed technologies, and the transition of these technologies from the research lab to an operational warehouse environment. A key les...