A method for solving the recognition of partially occluded parts is presented. It is based on the automatic generation of features from a set of primitive features which are configurations of pairs of fixed length segments of boundary edges of the parts. The procedure that creates the recognition features assigns a number in the range (0,1) that indicates the importance of the feature in the recognition strategy. This number is referred to as the feature's saliency. The method assumes that the parts that can occur in a scene come from a known set of parts. An example illustrates how automatically generated features can be used to count the number of identical parts in a heap.

An efficient on-line scheme for computing the inverse joint solution of robotic manipulators is combined with an improved formulation of robust, non-linear feedback control in joint space to produce a realizable Cartesian control scheme. Parametric uncertainties in the robot model are highlighted by the inclusion of compliance at each joint. Simulation results for a two link, coupled manipulator demonstrate that this Cartesian control enables the tip of the arm to track the demanded trajectory with arbitrarily small error in response to realistic actuator torques.

A direct subspace of a dynamic three-dimensional joint space is found to be useful for robot path planning in workspaces comprised of both static and dynamic objects. Dynamic descriptions permit positioning tables, automated guided vehicles, conveyors and cycling machine tools to be modeled by elements which translate or cycle along rectilinear paths. Graphical path planning procedures use cursor indicators to move the robot configuration point between the desired starting and final configurations while avoiding both the static and dynamic joint space obstacles.

Programming and control of industrial robots in a flexible manufacturing environment is a problem of increasing complexity. The paper deals with a software system, which is as much as possible independent of the real robot and suitable for different microprocessor configurations. It contains, furthermore, some components for off-line programming.

We present an analysis of currently available user language facilities for sensor-based programming of robot assembly tasks. A common criterion is defined that a language support for sensor-based programming must satisfy. Finally, drawing upon a novel mechanism of generalized exception, we present a framework for a complete solution of these problems in a high level block structured and complex language, as well as a possible implementation of such tools.

The need to store the trajectory information required by industrial robots, so that they may carry out tasks such as painting, has led to various data compression methods. The method proposed, simple to implement and adaptable to low cost systems, enables a best compromise to be reached for a given application between the detailed description of a complex movement and the use of as little memory as possible. This method was first used on a hydraulic prototype for the French Master-Slave firm PHAREMME.

The paper describes a study of potential industrial applications where modular robots can be used in preference to conventional pedestal mounted robots and/or specially designed automatic machines. The study was conducted with reference to a particular family of pneumatically actuated robot modules.

The paper also describes a modular robotic system which was designed to automate assembly processes in the manufacture of a family of egg coddlers. The manipulator comprises multi-axis groups of modules which operate concurrently to perform the assembly operations. Flexibility is incorporated through the use of a hierarchical control system architecture.

We suggest a new pretopological model for pattern recognition which was introduced to study complex economic systems. The model has its origin in the concept of “neighbour”, which is both primitive and fundamental in pattern recognition. Pretopology enables us to develop a perceptive and topological approach for patterns and to see that problems, apparently different, are in fact identical e.g. clustering and recognition, search of skeletons in image processing and search of an informative learning set. It should be noted that the suggeted model is more than a descriptive one.

One of the most important specifications associated with a robot manipulator is repeatability. Unfortunately, there are no standard procedures for determining or specifying the repeatability of a robot. This situation has resulted in a good deal of confusion when attempting to use a published repeatability specification to determine if a robot will meet the requirements of an application.

The purpose of this paper, therefore, is to propose a systematic procedure for the determination and specification of robot repeatability. To accomplish this, three aspects of repeatability analysis are examined. First, a method of collecting data to study repeatability is discussed. A data acquisition system is described as well as an associated test procedure. Next, a procedure for utilizing the acquired data to develop a repeatability specification for a robot is proposed. This procedure results in a specification of the robot's ability to orient objects as well as the classical specification for positioning repeatability. A methodical approach to determining where in the workspace the repeatability test should be made is also addressed. Finally, an approach is presented whereby the data collected during the repeatability test may be used to determine which joints contribute most to the variations in robot repeatability.

Path planning a robot arm motion essentially requires that the constraints of the joint variables and the vector of the joint motion rates are taken into account. In order to satisfy the constraints of the joint variables a sliding mode is being employed together with the developed kinematic path control method. The extended form of the kinematic path control method, here proposed, treats simultaneously the constraints of the joint variables and the vector of joint motion rates in path planning a robot arm motion.

The optimum control of an industrial robot can be achieved by splitting the problem into two tasks: off-line programming of an optimum path, followed by an on-line path tracking.

The aim of this paper is to address the numerical solution of the optimum path planning problem. Because of its mixed nature, it can be expressed either in terms of Cartesian coordinates or at joint level.

Whatever the approach adopted, the optimum path planning problem can be formulated as the problem of minimizing the overall time (taken as objective function) subject to behavior and side constraints arising from physical limitations and deviation error bounds. The paper proposes a very general optimization algorithm to solve this problem, which is based on the concept of mixed approximation.

A numerical application is presented which demonstrates the computational efficiency of the proposed algorithm.

Let P = {p1, …,pn} and Q = {q1,…,qm} be two simple polygons in the plane with non-intersecting interiors, the vertices of which are specified by their cartesian coordinates in order. The translation separability query asks whether there exists a direction in which P can be translated by an arbitrary distance without colliding with Q. It is shown that all directions that admit such a motion can be computed in O(nlogm) time, where n > m, thus improving the previous complexity of O(nm) established for this problem. In designing this algorithm a polygon partitioning technique is introduced that may find application in other geometric problems.

The algorithm presented in this paper solves a simplified version of the grasping problem in robotics. Given a description of a robot hand and a set of objects to be manipulated, the robot must determine which objects can be grasped. The solution given here assumes a two-dimensional world, a hand without an arm, and grasping under translation only.

The concepts of entity-relationship diagram have been applied to picture description. Primitive picture entities, picture relationships, and picture grammars are presented with illustrative examples. A high-level description of a two-level picture generation system is proposed using either string description or ER diagram description. Illustrative examples are also given. The advantages of ER diagram description together with its comparison to string description are also presented. The results may have useful applications in robotics, artificial intelligence, expert systems, picture processing, pattern recognition, knowledge engineering and pictorial database design.

This work describes a hierarchically structured geometric database in an off-line robot programming system. The data structure contains the numerical definition of the frame variables, as well as an indicator of the respective reference frames. Moreover, the physical relations between the objects in the environment are included. The database is implemented such that it continuously reflects the actual structure of the environment. As a result, all calculations of the frame locations are carried out automatically. Moreover, the programming system is capable to autonomously updating the numerical information after changes in the environment. Making this database the heart of a robot programming system greatly simplifies the off-line programming of complex robot tasks, like f.i. assembly tasks.

This paper presents current trends in robot programming. The open problems with current robot programming systems are outlined and indications for solutions are given. Since computer controlled robots have been introduced, the methodology of robot programming has seen a great deal of development. Two completely different approaches to robot programming have been considered in the past. On the one hand within the Artificial Intelligence community a lot of research has been done to provide robots with autonomous reasoning capabilities. On the other hand, the need to control industrial robots has pushed the development of simple but effective methods for robot programming. To put it simply, Artificial Intelligence researchers have taken a top-down approach trying to solve the difficult problem of reasoning and have assumed that all the rest was easy. Others have taken a bottom-up approach first trying to control robots and only later trying to incorporate intelligence. The complexity of industrial automation tasks requires programming systems more sophisticated that those in use today. Artificial Intelligence is the best candidate to create the next generation of robot programming systems.