The problem of dual-arm manipulation planning is to plan the path of two cooperating robot arms to carry a moveable object from a given initial configuration to a given goal configuration amidst obstacles. The authors...
详细信息
The problem of dual-arm manipulation planning is to plan the path of two cooperating robot arms to carry a moveable object from a given initial configuration to a given goal configuration amidst obstacles. The authors survey previous work in manipulation planning and relate it to this analysis. A formal configuration space presentation of the problem is given. Three implemented planners are described. These three planners operated in a two-dimensional workspace and addressed increasingly difficult versions of the dual-arm manipulation problem. They were implemented in the C language. Execution times given were obtained by running them on a DEC 5000 workstation.< >
The problem of partitioning an assembly of polyhedral objects into two subassemblies that can be separated arises in assembly planning. The authors describe an algorithm to compute the set of all translations separati...
详细信息
The problem of partitioning an assembly of polyhedral objects into two subassemblies that can be separated arises in assembly planning. The authors describe an algorithm to compute the set of all translations separating two polyhedra with n vertices in O(n/sup 4/) steps and show that this is optimal. Given an assembly of k polyhedra with a total of n vertices, an extension of this algorithm identifies a valid translation and removable subassembly in O(k/sup 2/n/sup 4/) steps if one exists. Based on the second algorithm, a polynomial time method for finding a complete assembly sequence consisting of single translations is derived. An implementation incorporates several changes to achieve better average-case performances. Experimental results obtained for composite objects consisting of isothetic polyhedra are described.< >
A method for converting a series of line segments generated by a high-level path planner for an AGV to a combination of continuous smooth curves and straight line paths is presented. Ideally, the straight line path se...
详细信息
A method for converting a series of line segments generated by a high-level path planner for an AGV to a combination of continuous smooth curves and straight line paths is presented. Ideally, the straight line path segments should be connected by circular arcs. The AGV's angular velocity along the straight paths is zero and will not smoothly connect the angular velocity moments required to traverse the circular arcs. Therefore, a transition curve (clothoid) needs to be generated to smoothly connect these angular moments. A technique for computing the clothoid transition curves for an AGV to execute is given. The method was implemented on a Macintosh II computer and was used to control a mobile robot.< >
Motivation for using harmonic functions as solutions to Laplace's equation arose from the resulting prevention of spurious local minima. Harmonic functions are shown to have several more useful properties that mak...
详细信息
Motivation for using harmonic functions as solutions to Laplace's equation arose from the resulting prevention of spurious local minima. Harmonic functions are shown to have several more useful properties that make them well suited for robotics applications such as fast surface normal computation, completeness, ability to exhibit different modes of behavior, and robust control in the presence of unanticipated obstacles and errors. A system that demonstrates this usefulness for control of a revolute robot arm is discussed.< >
A technique for the calibration of an active camera system is presented. The calibration of manipulator, camera-to-manipulator, camera, and base-to-world is treated in a unified and elegant way. In this approach, the ...
详细信息
A fitting relationship in an assembly implies that the relative location of the bodies belongs to a coset of the symmetry group of the mating feature pair. When a symmetry group is continuous, there are infinitesimal ...
详细信息
A fitting relationship in an assembly implies that the relative location of the bodies belongs to a coset of the symmetry group of the mating feature pair. When a symmetry group is continuous, there are infinitesimal displacements which preserve the relationship. Assembly of two bodies normally involves the establishment of successively more constraining relations, many of which are fitting relations. The continuous topological structure of the associated group determines possible directions of assembly at any state in the assembly process. To accommodate to errors, it is necessary to choose a stiffness matrix appropriate to a given assembly state, which allows the robot to comply with wrenches normal to the possible assembly directions. The derivation of such matrices from a computational geometric representation of the mating feature symmetry group is considered.< >
The rediscovery and popularization of the backpropagation training technique for multilayer perceptrons as well as the invention of the Boltzmann machine learning algorithm has given a new boost to the study on superv...
The rediscovery and popularization of the backpropagation training technique for multilayer perceptrons as well as the invention of the Boltzmann machine learning algorithm has given a new boost to the study on supervised learning networks. In recent years, besides widely spread applications and various further improvements of the classical backpropagation technique, many new supervised learning models, techniques as well as theories, have also been proposed in a vast number of publications. This paper tries to give a rather systematic review on the recent advances on supervised learning techniques and models for static feedforward networks. We summarize a great number of developments into four aspects: (1) Various improvements and variants made on the classical backpropagation techniques for Multilayer(static) perceptron nets, for speeding up training, avoiding local minima, increasing the generalization ability as well as for many other interesting purposes. (2) A number of other learning methods for training multilayer (static) perceptron, such as derivative estimation by perturbation, direct weight update by perturbation, genetic algorithms, recursive least square estimate and extended Kalman filters, linear programming, the policy of fixing one layer while updating another, constructing networks by converting decision tree classifiers and others. (3) Various other feedforward models which are also able to implement function approximation, probability density estimation and classification, including various models of basis function expansion (e.g. radial basis functions, restricted coulomb energy, multivariate adaptive regression splines, trigonometric and polynomial bases, projection pursuit, basis function tree and many others) and several other supervised learning models. (4) Models with complex structures, e.g. modular architecture, hierarchy architecture and others. Altogether, we try to give a global picture of the present state of supervised learning tech
暂无评论