Max Planck Lecture 2017
The field of transportation is undergoing a seismic change with the coming introduction of autonomous driving. The technologies required to enable computer driven cars involves the latest cutting edge artificial intelligence algorithms along three major thrusts: Sensing, Planning and Mapping.
NVIDIA CEO Jensen Huang presented the NVAIL AI Labs with the very first Tesla V100 GPUs, based on NVIDIA's Volta architecture. MPI-IS is among the top centers working at the leading edge of deep learning in computer vision. As such it is recognized by NVIDIA as one of its NVAIL labs and giving the MPI access to the best and latest NVIDIA technology. Huang unveiled these new GPUs at CVPR saying that he wants to put them in the hands of researchers first.
Researchers at the Max Planck Institute for Intelligent Systems (MPI-IS) have developed technology to digitally capture clothing on moving people, turn it into a 3D digital form, and dress virtual avatars with it. This new technology makes virtual clothing try-on practical.
The Max Planck Institute for Intelligent Systems is celebrating the opening of its new building in Tübingen
Tübingen, 12. Juli 2017. This opening ceremony involved teamwork between humans and machines. While the robot Apollo held the red ribbon, Minister-President Winfried Kretschmann and Martin Stratmann, President of the Max Planck Society, cut the symbolic cordon to officially open the Max Planck Institute for Intelligent Systems’ new building in Tübingen. They were assisted by Science Minister Theresia Bauer and Stefan Schaal, the Institute’s Managing Director. After a two-and-a-half-year construction period, the scientists recently moved into the Institute building made possible by funding worth millions from the federal state government. They are now conducting basic research on artificial intelligence here.
For the fifth time, the MLSS takes place in Tübingen
Cédric de Crousaz and Julian Viereck receive the ETH Medal for their outstanding Master Theses
at the 2017 IEEE/RAS International Conference on Robotics and Automation
The paper "Probabilistic Articulated Real-Time Tracking for Robot Manipulation" by Cristina Garcia Cifuentes, Jan Issac, Manuel Wüthrich, Stefan Schaal and Jeannette Bohg was finalist for the Best Robotic Vision paper at the 2017 IEEE/RAS International Conference on Robotics and Automation.
Robust and real-time Bayesian articulated object tracking methods, implemented in C++ and CUDA.
We release open-source code and data sets on Bayesian articulated object tracking. The library contains approaches towards problems ranging from single object tracking to full robot arm pose estimation. The data sets allow the quantitative evaluation of alternative approaches thanks to accurate ground-truth annotations.
An elastic membrane covered with tiny fibres paired with a pressure differential enables a new soft gripper system with a high adhesion performance even on curved surfaces
Robots generally need a gripper that adapts to three-dimensional surfaces. Such a gripper needs to be soft to adapt to a great variety of geometries, but not too soft, as it will detach easily and not be able to bear weight for very long. Researchers working with Metin Sitti at the Max Planck Institute for Intelligent Systems in Stuttgart developed a membrane equipped with microscopic fibres inspired by the fine hairs on a gecko's foot and attached it to a suction cup-like flexible body. An internal pressure differential ensures perfect conformation of the flexible gripper to a wide variety of surfaces and equally distributes the load over the entire contact interface. As a result, the researchers suppressed load induced stress concentrations at the edges, which strongly reduced the adhesion. The gripper demonstrates a 14-times higher adhesion than grippers without this load sharing mechanism.