Unified Approach to Control of Motions of Mobile Robots

An updated configuration-control scheme incorporates a distinction between holonomic and nonholonomic constraints. An improved computationally efficient scheme has been developed for on-line coordinated control of both manipulation and mobility of robots that include manipulator arms mounted on mobile bases. The present scheme is similar to the one described in "Coordinated Control of Mobile Robotic Manipulators." Both schemes are based on the configuration-control formalism. The major difference between the two schemes is that unlike the previous one, the present one incorporates an explicit distinction between holonomic and nonholonomic constraints (see figure). Both schemes are characterized by computational efficiency and flexibility, which are advantageous for on-line, real-time control.

In addition to the article cited above, several other prior articles in NASA Tech Briefs have discussed aspects of the configuration-control formalism. These include "Increasing the Dexterity of Redundant Robots" (NPO-17801), Vol. 14, No. 10 (October, 1990), page 88; "Redundant Robot Can Avoid Obstacles" (NPO-17852, Vol. 15, No. 10 (October, 1991), page 86; "Configuration-Control Scheme Copes with Singularities" (NPO18556), Vol 17, No. 2 (February, 1993), page 81; and "More Uses for Configuration Control of Robots" (NPO18607/NPO-18608), Vol. 17, No. 10 (October, 1993), page 120. To recapitulate: A robot has n degrees of freedom. The basic task is to make the end effector (the hand at the tip of the manipulator amm) follow a prescribed trajectory in m-dimensional coordinates (where m The additional task can be specified by the user and can include (but is not limited to) reaching around obstacles, avoiding collisions with objects in the workspace, maintaining one or more links of the manipulator arm in a desired pose, and/or optimizing the overall kinematics in both the redundant and nonredundant degrees of freedom to enhance overall manipulability. The additional task is mathematically modeled by a set of kinematic functions that, in effect, define the trajectory in the redundant degrees of freedom. The user can assign weighting factors to individual degrees of mobility or manipulation as well as to each task specification. The user can also change task specifications and weighting factors during operation.

In the present scheme, the degrees of mobility and the degrees of manipulation are treated within a common theoretically framework; to put it in slightly different terms, the mobile base and the manipulator are treated as closely interacting subsystems of the overall robotic system, rather than as two separate entities. Within this framework, the kinematic constraints upon the manipulator subsystem are holonomic, whereas those on the mobile base can be nonholonomic, depending on the type of mobile base. All degrees of freedom are treated on an equal footing according to the computationally efficient configuration-control formalism. The nonholonomic kinematic constraints (if any) fit naturally into the configuration-control formalism: the nonholonomic kinematic constraints, the desired motion of the end effector, and the additional task specified by the user are combined to form a set of augmented tasks. These tasks are then accomplished in a coordinated manner by use of the configuration-control equations to determine the motion in each mobility and manipulation degree of freedom.

Point of Contact:
Homayoun Seraji,
Mail Stop 198-219
Jet Propulsion Laboratory
4800 Oak Grove Drive
Pasadena, CA 91109
seraji@telerobotics.jpl.nasa.gov

Telerobotics Program page

Please email the site webmaster with any comments, criticisms or corrections for this page.
Maintained by: Dave Lavery
Last updated: May 10, 1996