Search NASA⌕ Search

Engineering topics

Barker, L. K.

Publications and source records attributed to Barker, L. K..

At least 19 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS1987

Coordination of multiple robot arms

Kinematic resolved-rate control from one robot arm is extended to the coordinated control of multiple robot arms in the movement of an object. The structure supports the general movement of one axis system (moving reference frame) with respect to another axis system (control reference frame) by one or more robot arms. The grippers of the robot arms do not have to be parallel or at any pre-disposed positions on the object. For multiarm control, the operator chooses the same moving and control reference frames for each of the robot arms. Consequently, each arm then moves as though it were carrying out the commanded motions by itself.

Barker, L. K.↗
NASA NTRS1986

Modified Denavit-Hartenberg parameters for better location of joint axis systems in robot arms

The Denavit-Hartenberg parameters define the relative location of successive joint axis systems in a robot arm. A recent justifiable criticism is that one of these parameters becomes extremely large when two successive joints have near-parallel rotational axes. Geometrically, this parameter then locates a joint axis system at an excessive distance from the robot arm and, computationally, leads to an ill-conditioned transformation matrix. In this paper, a simple modification (which results from constraining a transverse vector between successive joint rotational axes to be normal to one of the rotational axes, instead of both) overcomes this criticism and favorably locates the joint axis system. An example is given for near-parallel rotational axes of the elbow and shoulder joints in a robot arm. The regular and modified parameters are extracted by an algebraic method with simulated measurement data. Unlike the modified parameters, extracted values of the regular parameters are very sensitive to measurement accuracy.

Barker, L. K.↗
NASA NTRS1986

Overcoming Robot-Arm Joint Singularities

Kinematic equations allow arm to pass smoothly through singular region. Report discusses mathematical singularities in equations of robotarm control. Operator commands robot arm to move in direction relative to its own axis system by specifying velocity in that direction. Velocity command then resolved into individual-joint rotational velocities in robot arm to effect motion. However, usual resolved-rate equations become singular when robot arm is straightened.

Barker, L. K.↗
NASA NTRS1986

Theoretical three-and four-axis gimbal robot wrists

In high-performance flight simulations, a four-axis gimbal system allows all possible rotations with acceptable gimbal angle rates while it avoids the so-callled 'gimbal lock' that occurs when gimbal rotational axes are colinear. In this paper, pertinent equations (including quaternions) are assembled for a hypothetical robot wrist, functionally equivalent to this four-axis gimbal system, and also for a true three-axis gimbal robot wrist. These equations are used to simulate the rotation of a robot hand by the robot wrist in response to operator rotational velocity commands to the robot hand. Near gimbal lock (wrist singularity), excessive rotational rates occur. Scaling the rates, which is necessary for the three-gimbal robot wrist to prevent rate limiting, introduces an undesirable time delay in the robot hand rotation with respect to the commanded rotation. However, the merit of the four-gimbal robot wrist is that the fourth gimbal angle keeps the robot wrist away from the singularity so that the robot hand moves exactly as commanded. It appears that in a 'worst-type' maneuver of the robot hand, the fourth gimbal angle can be defined so that none of the gimbal angle rates exceed about twice the commanded rates.

Barker, L. K.↗
NASA NTRS1986

Computing Relative Joint Positions of Robot Arms

Vector-algebra method developed for extracting Denavit-Hartenberg parameters for any assembled robot arm. Method for extracting relative joint geometry of robot arms useful to researchers who need data for existing robot arms for either validation of mathematical models or for studies involving actual control of these devices. Method, does not require robot arm to be disassembled, also useful in recalibration of misalined or bent robot arm and becomes useful industrial procedure. Merit of method is errors not propagated.

Barker, L. K.↗
NASA NTRS1984

Translational control of a graphically simulated robot arm by kinematic rate equations that overcome elbow joint singularity

An operator commands a robot hand to move in a certain direction relative to its own axis system by specifying a velocity in that direction. This velocity command is then resolved into individual joint rotational velocities in the robot arm to effect the motion. However, the usual resolved-rate equations become singular when the robot arm is straightened. To overcome this elbow joint singularity, equations were developed which allow continued translational control of the robot hand even though the robot arm is (or is nearly) fully extended. A feature of the equations near full arm extension is that an operator simply extends and retracts the robot arm to reverse the direction of the elbow bend (difficult maneuver for the usual resolved-rate equations). Results show successful movement of a graphically simulated robot arm.

Barker, L. K.↗
NASA NTRS1984

Kinematic control of robot with degenerate wrist

Kinematic resolved rate equations allow an operator with visual feedback to dynamically control a robot hand. When the robot wrist is degenerate, the computed joint angle rates exceed operational limits, and unwanted hand movements can result. The generalized matrix inverse solution can also produce unwanted responses. A method is introduced to control the robot hand in the region of the degenerate robot wrist. The method uses a coordinated movement of the first and third joints of the robot wrist to locate the second wrist joint axis for movement of the robot hand in the commanded direction. The method does not entail infinite joint angle rates.

Barker, L. K.↗
NASA NTRS1983

Kinematic equations for resolved-rate control of an industrial robot arm

An operator can use kinematic, resolved-rate equations to dynamically control a robot arm by watching its response to commanded inputs. Known resolved-rate equations for the control of a particular six-degree-of-freedom industrial robot arm and proceeds to simplify the equations for faster computations are derived. Methods for controlling the robot arm in regions which normally cause mathematical singularities in the resolved-rate equations are discussed.

Barker, L. K.↗
NASA NTRS1983

Vector-algebra approach to extract Denavit-Hartenberg parameters of assembled robot arms

The Denavit-Hartenberg parameters characterize the joint axis systems in a robot arm and, naturally, appear in the transformation matrices from one joint axis system to another. These parameters are needed in the control of robot arms and in the passage of sensor information along the arm. This paper presents a vector algebra method to determine these parameters for any assembled robot arm. The idea is to measure the location of the robot hand (or extension) for different joint angles and then use these measurements to calculate the parameters.

Barker, L. K.↗
NASA NTRS1983

Theoretical method for calculating relative joint geometry of assembled robot arms

Equations are developed to extract the relative joint parameters of an assembled robot arm. Specifically, the Denavit-Hartenberg parameters, which completely characterize the relative joint geometry, are calculated. These parameters are needed to control the hand of the robot arm by resolved rate. As an example, the parameter extraction equations are used with perfect simulated data (no measurement noise) obtained from a mathematical model of a six-degree-of-freedom robot arm. For an actual application, measurement data needed to estimate the relative joint parameters can be generated by moving a robot arm to different positions, measuring the location of the hand (or other extension) in base coordinates, and recording the corresponding joint angles.

Barker, L. K.↗
NASA NTRS1983

Decision-making and problem-solving methods in automation technology

The state of the art in the automation of decision making and problem solving is reviewed. The information upon which the report is based was derived from literature searches, visits to university and government laboratories performing basic research in the area, and a 1980 Langley Research Center sponsored conferences on the subject. It is the contention of the authors that the technology in this area is being generated by research primarily in the three disciplines of Artificial Intelligence, Control Theory, and Operations Research. Under the assumption that the state of the art in decision making and problem solving is reflected in the problems being solved, specific problems and methods of their solution are often discussed to elucidate particular aspects of the subject. Synopses of the following major topic areas comprise most of the report: (1) detection and recognition; (2) planning; and scheduling; (3) learning; (4) theorem proving; (5) distributed systems; (6) knowledge bases; (7) search; (8) heuristics; and (9) evolutionary programming.

Hankins, W. W.↗
NASA NTRS1981

Marking parts to aid robot vision

The premarking of parts for subsequent identification by a robot vision system appears to be beneficial as an aid in the automation of certain tasks such as construction in space. A simple, color coded marking system is presented which allows a computer vision system to locate an object, calculate its orientation, and determine its identity. Such a system has the potential to operate accurately, and because the computer shape analysis problem has been simplified, it has the ability to operate in real time.

Bales, J. W.↗
NASA NTRS1980

Stability boundaries for systems with frequency-model feedback and complacency filter

General parameter-plane equations were derived to generate boundaries for a class of systems characterized by a feedback loop that contains a complementary filter and a model for either the low- or high-frequency portion of the plant. This combination allows those frequencies of the part of the plant that is modeled to be fed back for control while suppressing other frequencies. For all specific examples considered, the stability regions obtained using the complementary filter and frequency model were larger (and in some cases, considerably larger) than those obtained using a low pass filter in the feedback of the system output. Furthermore, higher gain control was possible.

Barker, L. K.↗
NASA NTRS1979

Mikhailov Stability Criterion for Time-delayed Systems

The valid and invalid application of the Mikhailov criterion to linear, time-invariant systems with time delays is discussed. The Mikhailov criterion is a graphical procedure which was developed to examine the stability of linear, time-invariant systems with no time delays. Two equivalent formulations of the criterion are discussed. Results indicate that the first formulation remains valid for time-delayed systems of the retared type, with the understanding that the Mikhailov curve need not necessarily always rotate in the counterclockwise direction for a stable system. Erroneous results in the second formulation are formed when there are time delays in the systems.

Barker, L. K.↗
NASA NTRS1977

Stability analysis for linear systems with time delays

Time delays in the mathematical description of a physical system occur whenever the system is affected not only by conditions at the present time, but also by conditions which have occurred in the past. The tau-decomposition method, as refined by Lee and Hsu, is a method for studying the effects of time delay on the stability of retarded dynamical systems. In this paper, the method is extended to examine the stability and relative stability of: (1) retarded systems with many time delays; and (2) a class of neutral systems with one delay. Applications are included which illustrate the power of the technique.

Barker, L. K.↗
NASA NTRS1976

Stability of neutral equations with constant time delays

A method was developed for determining the stability of a scalar neutral equation with constant coefficients and constant time delays. A neutral equation is basically a differential equation in which the highest derivative appears both with and without a time delay. Time delays may appear also in the lower derivatives or the independent variable itself. The method is easily implemented, and an illustrative example is presented.

Barker, L. K.↗
NASA NTRS1975

Analytical method for determining the stability of linear retarded systems with two delays

The stability is considered of the solution differential-difference equations of the retarded type with constant coefficients and two constant time delays. A method that makes use of analytical expressions to determine stability boundaries, and the stability of the system, is derived. The method was applied to a system represented by a second-order differential equation with constant coefficients and time delays in the velocity and displacement terms. The results obtained is in agreement with those obtained by other investigators.

Barker, L. K.↗