Time-optimal extension and retraction of robots: Numerical analysis of the switching structure

The problem of the time-optimal control of robot manipulators is of importance because of its potential for increasing the productivity of assembly lines. This work is part of a series of papers by the authors on this topic using direct and indirect methods of optimization. A cylindrical robot or a spherical polar robot constrained to the horizontal plane is considered, and optimal solutions for radial maneuvers are generated. Indirect methods are employed in order to establish the switching structure of the solutions. The results show that even such apparently simple maneuvers as extension or retraction of a robot with a prismatic joint can produce very complex optimal solutions. Time-optimal retraction can exhibit ten different switching structures with eight switching points and two singular arcs.This research was supported by the Deutsche Forschungsgemeinschaft. The results reported in this paper first appeared in the thesis of the first author and were presented at the 1989 AIAA Guidance, Navigation and Control Conference (Refs. 1 and 2).     

Study of the energy flow in sulphur-and oxygen-nucleus collisions at 60 and 200 GeV/nucleon

Data on multiplicities of charged particles produced in proton-nucleus and nucleus-nucleus collisions at 200 GeV per nucleon are presented. It is shown that the mean multiplicity of negative particles is proportional to the mean number of nucleons participating in the collision both for nucleus-nucleus and proton-nucleus collisions. The apparent consistency of pion multiplicity data with the assumption of an incoherent superposition of nucleon-nucleon collisions is critically discussed.     

Centroaffine Bernstein problems

C.P. Wang [Geom. Dedicata 51 (1994) 63–74] studied the Euler–Lagrange equation for the centroaffine area functional of hypersurfaces. In terms of a local representation of the hypersurface as a graph, this equation is a complicated, strongly non-linear fourth order PDE. We consider classes of solutions satisfying these equations together with completeness conditions. We also formulate appropriate centroaffine Bernstein problems and give partial solutions.     

Coated film electrodes

Summary The construction of coated-film electrodes is described. It is shown that they are miniaturizable, inexpensive, simple sensors for various cations and anions. They are similar to the coated-wire electrodes, but instead of a metal wire an extremely thin conductive metal layer manufactured in thin-film technology is used. Potassium and sodium electrodes studied so far show excellent electrode properties, i. e. near Nernst response over a wide concentration range, sometimes better than commercial ion-selective electrodes. Two film electrodes are used to develop a low cost sensor for the determination of the activity of ions in a simple way by potentiometric difference measurement; this solid-state sensor can be applied to biomedical measurements, especially for blood analysis.

---

Filmelektroden

---

Dedicated to Prof. Dr. G. Tölg on the occasion of his 60th birthday     

A perspective on artificial intelligence: Learning to learn

The classical approach to the acquisition of knowledge in artificial intelligence has been to program the intelligence into the machine in the form of specific rules for the application of the knowledge: expert systems. Unfortunately, the amount of time and resources required to program an expert system with sufficient knowledge for non-trivial problem-solving is prohibitively large. An alternative approach is to allow the machine tolearn the rules based upon trial-and-error interaction with the environment, much as humans do. This will require endowing the machine with a sophisticated set of sensors for the perception of the external world, the ability to generate trial actions based upon this perceived information, and a dynamic evaluation policy to allow it to measure the effectiveness of its trial actions and modify its repertoire accordingly. The principles underlying this paradigm, known ascollective learning systems theory, have already been applied to sophisticated gaming problems, demonstrating robust learning and dynamic adaptivity.The fundamental building block of a collective learning system is thelearning cell, which may be embedded in a massively parallel, hierarchical data communications network. Such a network comprising 100 million learning cells will approach the intelligence capacity of the human cortex. In the not-too-distant future, it may be possible to build a race of robotic slaves to perform a wide variety of tasks in our culture. This goal, while irresistibly attractive, is most certainly fraught with severe social, political, moral, and economic difficulties.This paper was given as an invited talk on the 12th Symposium on Operations Research, University of Passau, September 1987.