Keynote Speakers

Abstract

Computations in robotics cover a rich spectrum of problems at the junction of mechanics, computer science, engineering and mathematics. In this talk the emphasis will be given to robot kinematics, its analysis, design and optimization with respect to different tasks. This will include a brief discussion on the direct and inverse kinematics problems of serial and parallel mechanisms, kinematic singularities, workspace determination, manipulability, as well as kinematic flexibility. Examples will present peculiarities in robot and human motion performing different tasks, such as the manipulation of heavy objects or the vertical jump, with the focus on kinematic redundancy. Redundant robots possess too many degrees of freedom; their number exceeds the number required by the task. Therefore, redundancy gives to the robot an immense source of freedom and enables it to solve different tasks in an infinite number of ways. The robot can simultaneously solve additional secondary tasks of lower priority. We will discuss various aspects in robot programming and design in comparison with humans.

Abstract

Many industries use simulation tools for virtual product design, and there is a growing trend towards using simulation in combination with optimization algorithms. The requirements for optimization under such circumstances are often very strong, involving many design variables and constraints and a strict limitation of the number of function evaluations to a surprisingly small number (often around one thousand or less). Tuning optimization algorithms for such challenges has led to very good results obtained by variants of evolution strategies. Evolutionary algorithms are nowadays standard solvers for such applications. In the presentation, sample cases from industry are presented, their challenges are discussed in more detail. Results of an experimental comparison of contemporary evolution strategies on the BBOB test function set for a small number of function evaluations are presented and discussed, and further enhancements of contemporary evolution strategies are outlined. Our practical examples are motivated by industrial applications. A typical challenge is to find innovative solutions to a design optimization task. Based on a suitable definition of innovative solutions, an application of this concept to an airfoil design optimization task is discussed in the presentation. Discussing these applications and the variants of evolution strategies applied, the capabilities of these algorithms for optimization cases with a small number of function evaluations are illustrated.

Abstract

Evolutionary Computing (EC) is the research field concerned with artificial evolutionary processes in digital spaces, inside computers. In about three decades the EC community learned the ‘art of taming evolution’ and developed several evolutionary algorithm variants to solve optimization, design, and machine learning problems. In all these applications, the reproductive entities are digital. This holds even in evolutionary robotics where the evolved code is ported to a robot body and in evolutionary design where the evolved design is constructed physically after the evolutionary process terminates. In this talk I present a vision about the next big breakthrough: the creation of artificial evolutionary processes in physical spaces. In other words, I envision the ``Evolution of Things'', rather than just the evolution of digital objects, leading to a new field of Embodied Artificial Evolution. After presenting this vision I elaborate on some of the technical challenges and relate the main algorithmic/technical requirements to the current know-how in EC. Finally, I will speculate about possible applications, their societal impacts, and argue that these developments will radically change our lives.