Data processing: generic control systems or specific application – Specific application – apparatus or process – Robot control
Reexamination Certificate
2002-07-23
2004-02-10
Cuchlinski, Jr., William A. (Department: 3661)
Data processing: generic control systems or specific application
Specific application, apparatus or process
Robot control
C700S246000, C700S247000, C700S248000, C700S249000, C700S250000, C700S251000, C700S252000, C700S253000, C700S260000, C700S261000, C700S262000, C700S263000, C700S264000, C318S568100, C318S568110, C318S568120, C318S568220, C901S001000, C901S033000, C901S045000, C901S046000, C901S047000, C901S048000, C701S023000, C701S301000, C600S595000
Reexamination Certificate
active
06690999
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method for controlling an industrial manipulator used for assembly, painting and operations in a factory, or a space manipulator used for carrying out inspections and assembly operations and the like in space. The invention particularly relates to a method for effectively controlling a hyper-redundant type manipulator having the degrees of freedom exceeding the required degrees of freedom.
2. Description of the Prior Art
In a manipulator in which a plurality of links are connected by joints, the necessary angle that each joint must have in order to move the manipulator tip to a target position is generally calculated by making up a Jacobi matrix for calculating tip displacement from the joint angle displacement, and using an inverse Jacobi matrix corresponding to the inverse transformation of the Jacobi matrix to find the joint angle values.
An inverse Jacobi matrix can be uniquely obtained with respect to a manipulator in which the degrees of freedom are the same as the required degrees of freedom. However, in the case of hyper-redundant manipulators that have more than the necessary degrees of freedom, it cannot be uniquely obtained because the Jacobi matrix becomes a rectangular matrix. Therefore, what designers have been doing is setting a number of constraint conditions and using singular value analysis or Lagrange's undetermined multiplier method or the like to calculate a pseudo inverse matrix.
However, when the above method is used to control hyper-redundant manipulators, the problem has been that as the difference between the necessary degrees of freedom and the degrees of freedom possessed by the manipulator increases, that is, as the degree of redundancy increases, the calculations become increasingly complex. In addition, it has been necessary to use special calculations to adapt to partial failure of the manipulator. It is also necessary to perform these calculations in a centralized manner, which has not been suitable for the distributed control structures that have been coming into general use in recent years.
Considerable work is underway to develop hyper-redundant manipulators that can move flexibly, avoid obstacles and cope with partial failures. In particular, manipulators with high hyper redundancy are being developed that when used for inspecting satellites in orbit, are adept at getting around objects and can continue to function even after suffering a partial failure. When this type of space-based application is considered, the calculation resources available for controlling the manipulator are highly limited by the harshness of the environmental conditions, such as radiation and extremely low temperatures. Therefore, there is a limit when it comes to centralized computation of joint angle values in the case of a manipulator having a high hyper redundancy. Development is therefore focusing on getting around individual constraints on computing resources by using processors to control joints, a modular configuration and decentralized control to achieve robustness with respect to failures. This being the case, it is also desirable to be able to readily apply such a distributed control structure as a way of determining manipulator joint angles.
An object of the present invention is to provide a manipulator control method that employs a technique of determining angles of manipulator joints to control a manipulator using a distributed control structure, that can be readily applied to manipulators having a high degree of redundancy and to manipulators that suffer failure.
SUMMARY OF THE INVENTION
To attain the above object, the present invention provides a method for controlling a hyper-redundant manipulator comprising a plurality of links coupled by joints, by determining a shape the manipulator forms from base to end when the end is moved to a target position, the method comprising: modeling each link as an elastic body having a natural length and an appropriate modulus of elasticity that enables the elastic body to stretch and contract just in a linkage direction, simulating an overall shape of the manipulator when the end has been moved to a target position with the joints locked at a freezed angle and the joints are unlocked to return each link to its natural length, and moving the manipulator end to the target position by controlling each joint angle to match the simulation outcome.
In the above manipulator control method, the modulus of elasticity of the elastic body constituting each link has a nonlinear characteristic.
In the method, when there is an obstacle to the movement of the manipulator end to the target position, the overall shape of the manipulator is determined by introducing the obstacle as an external force acting on each elastic body that is inversely proportional to a distance between the obstacle and each elastic body.
In the method, when a joint is unable to function, the overall shape of the manipulator is determined by connecting links on each side of the failed joint to assume a virtual link that is introduced as an assumed elastic body that can stretch and contract along a straight line that connects the ends of the virtual link and is treated as the natural length of the virtual link when the joint has become unmovable.
In the method, when the manipulator is selectively moved to a target position, the overall shape of the manipulator is determined by using a plurality of processing means to carry out a decentralized simulation of the overall shape.
With the manipulator control method of this invention, the time-evolution of a spring with a given displacement can be found by calculating the displacement produced in the spring, facilitating the calculation and making the method applicable to hyper-redundant manipulators. The time-evolution of a spring with a given displacement can also be calculated from the spring displacement and the external force at each end of the spring, making it possible to realize manipulator control using a decentralized control structure. The method can also be applied to a faulty manipulator by introducing a virtual link that includes the faulty joint.
Further features of the invention, its nature and various advantages will be more apparent from the accompanying drawings and following detailed description of the invention.
REFERENCES:
patent: 4872665 (1989-10-01), Chareire
patent: 5055755 (1991-10-01), Ozawa et al.
patent: 5184319 (1993-02-01), Kramer
patent: 6059506 (2000-05-01), Kramer
patent: 6181983 (2001-01-01), Schlemmer et al.
patent: 6243623 (2001-06-01), Takenaka et al.
patent: 02003039353 (2003-02-01), None
Communications Research Laboratory, Independent Administrative I
Cuchlinski Jr. William A.
Marc McDieunel
Oblon & Spivak, McClelland, Maier & Neustadt P.C.
LandOfFree
Manipulator control method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Manipulator control method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Manipulator control method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3295230