内容简介：Described  in this lecture are basic principles for formulation of continuous  differential-geometric approaches which can be of use for fast kinematics and  dynamics computations. In the introduction, the concept of kinetostatic  transmission element is briefly reviewed, representing a mechanical  interpretation of generic differential-geometric mappings between manifolds. In  a first part, the solution of rigid-body motion along a SE(3) track using  quintic splines for continuous position and rotation interpolation is described   – termed the “Spline Joint”. The type of interpolation – Frenet, Darboux,  Bishop, Quaternion – can be regarded as an internal detail and “hidden” from the  user, allowing for easy representation switches and their combination. The  application as a layout program for roller coasters shows its industrial  relevance. The concept is then generalized to waterslides by regarding a  corresponding “Surface Joint” with a Darboux frame continuously sliding on the  surface and computing the stationary water channel through Bernoulli’s model and  its interaction with the sliding person through water mass rate resistance. The  approach shows that the stationary water channel as well as the person can be  predicted highly accurately and by a factor of 1:630,000 faster than commercial  fluid mechanics codes. In the second part, the kinematics of shortest paths on  smooth frictionless surfaces is regarded, as can be of use e.g. to compute  musculotendon systems but also for soft finger actuators. The approach starts by  regarding exact geodesics on surfaces as closing conditions, and then by  regarding also “thick” strands with varying cross section wrapped around a  surface. Its is shown again that, while the derivations and the differential  geometry are non-trivial, the final code is easy to implement and faster than  discrete methods by factors of up to 100. In the final part, the use of motion  and force increment screws and the property of eigenvectors as convergence of  power iterations of the impedance matrix to the eigenvalue with the highest  impedance (=the lowest stiffness) allows one to identify an unknown joint axis  by only application of force-displacement iterations, as shown from an ongoing  research project RobDIP together with a chief orthopedic surgeon and industrial  robot company.

报告人简介：Professor  Kecskeméthy  was born in Lima, Peru and graduated from the University of Stuttgart in  Mechanical Engineering in 1984, receiving his Ph.D. in Mechanical Engineering  from the University of Duisburg in 1993. From 1994 to 1995, he stayed as a  senior guest researcher at the Centre for Intelligent Machines at McGill  University with a fellowship from the German Research Foundation (DFG). In 1996,  Dr. Kecskemethy was appointed Professor at the Technical University of Graz,  where he held the Chair for Mechanics until 2002. In 2002, Professor Kecskeméthy  moved to the University of Duisburg-Essen, where he holds the Chair for  Mechanics and Robotics at the Institute of Mechatronics and System Dynamics. He  was Editor-In-Chief of the Journal Mechanism and Machine Theory from 2004 to  2015. He served as Dean of the Faculty of Engineering and as Chairman of the  Senate of the University Duisburg-Essen. His memberships include the German  Engineering Association VDI and the International Federation for the Promotion  of Mechanism and Machine Science IFToMM. Since 2013, he was IFToMM Germany  Chair, and since 2020 he is IFToMM President for the term 2020-2023. Professor  Kecskeméthy has worked in the areas of kinematics and dynamics of multibody  systems, covering topics such as modelling and control of mechatronic systems,  vehicle dynamics, design of legged machines, biomechanics, virtual reality as  well as medical and heavy-weight robotics. He has organized several  international conferences [IAK 2008, IAK 2013, IAK 2018, Computational  Kinematics 2009, ECCOMAS Multibody Dynamics 2019, International Symposium on  Theory and Practice of Robots and Manipulators (ROMANSY) 2022, Advances in Robot  Kinematics (ARK) 2022], has been Editor-In-Chief of the Journal Mechanism and  Machine Theory from 2004 to 2015, and is the author of more than 230 proceedings  and journal papers.