- Annette Beyer (for administrative matters)
- Schloss Dagstuhl: Zukünftige Karten für Menschen und Maschinen
Wolfgang Back und Wolfgang Rudolph im Gespräch mit Dr.-Ing. Claus Brenner (Interview ab 10:32)
In recent years, the advent of car navigation systems has laid the ground for an entirely new industry sector, consisting of map producers, car/ personal/ smart phone navigation manufacturers, and service providers. It has probably gone unnoticed that navigation systems mark a major change in the way we use maps. Partially, they are still just a replacement for traditional maps, providing a means to store and visualize a representation of the environment. In contrast to the traditional use of maps, however, navigation systems perform computations using the map’s data structures, such as shortest route, map matching, and route guidance. That is, from an abstract point of view, part of the map is made for machine use only – the user has no direct access to it but rather is only presented the outcome of the computations.
This development will accelerate in the near future, as sensor, processing, and communication capabilities of cars, or in general, robots, increase. Maps will evolve from their traditional meaning towards virtual representations of the environment, containing information specifically tailored for the employed sensors, algorithms, and questions to be solved. For example, a map may contain geo-referenced feature descriptors which are recognized by a given algorithm in the live video stream of a car’s camera, allowing to solve the specific problem of highly accurate positioning.
This development will accelerate in the near future, as sensor, processing, and communication capabilities of cars, or in general, robots, increase. Maps will evolve from their traditional meaning towards virtual representations of the environment, containing information specifically tailored to the employed sensors, algorithms, and questions to be solved. For example, a map may contain high¬dimensional local descriptors which can be recognized by a given algorithm in the live video stream of a car’s camera, allowing solving the positioning problem with unprecedented accuracy and reliability.
Such maps will not only be very different from today’s maps, they also will have to be produced by entirely different mechanisms. Since they contain very many details, it is infeasible to use (partially) manual processes to produce them, as is done today. Also, the mapping of details usually implies a high change rate, calling for frequent reacquisition, which is not economically feasible using traditional surveying and mapping techniques. What is required is an approach which is both, inexpensive and accurate, detailed and up to date, using a hierarchy of measurement systems which include the map users themselves.
The purpose of the ‘dynamic maps’ seminar was to bring together researchers from academia and industry, from several fields of computer science, including computer vision, photogrammetry, robotics, computer graphics, geoinformatics, and driver assistance. Central objectives of the seminar were to alert the different communities to the efforts in the respective other communities, to exchange our ideas about such a future, ‘dynamic’ map, and to foster future research proposals in this area.
The organizers thank all participants for their lively presentations and the fruitful discussions. We think that the seminar was a great success and served its main purpose, namely, making the different communities aware of each other, to a great extent.
It is clear that ‘dynamic maps’ is, and will continue to be, an active research topic for many years to come. Several attendants of the seminar have applied for research grants, as part of a larger German Research Foundation bundle project, in December 2010. It is worth to note that Wolfram Burgard has been granted an ERC advanced grant in 2010 for the topic of ‘LifeNav – Reliable Lifelong Navigation for Mobile Robots’, confirming the importance of the topic.
It is also clear that this area is so wide and complex, involving sensors, vision algorithms, in‐car systems, acquisition strategies, data modeling, storage, and serving, that it will require a combined effort of all players to move research forward, and finally develop systems that will ultimately change the lives of all of us. That said, the organizers are looking forward to further seminars on the topic of ‘dynamic maps’.
- Sven Behnke (Universität Bonn, DE) [dblp]
- Maren Bennewitz (Universität Freiburg, DE) [dblp]
- Dorit Borrmann (Jacobs Universität - Bremen, DE) [dblp]
- Claus Brenner (Leibniz Universität Hannover, DE) [dblp]
- Wolfram Burgard (Universität Freiburg, DE) [dblp]
- Jan Effertz (Volkswagen AG - Wolfsburg, DE)
- Jan-Michael Frahm (University of North Carolina at Chapel Hill, US) [dblp]
- Uwe Franke (Daimler AG - Sindelfingen, DE) [dblp]
- Friedrich Fraundorfer (ETH Zürich, CH) [dblp]
- Udo Frese (Universität Bremen, DE)
- Simone Frintrop (Universität Bonn, DE)
- Giorgio Grisetti (Universität Freiburg, DE)
- Jan-Henrik Haunert (Universität Würzburg, DE) [dblp]
- Olaf Hellwich (TU Berlin, DE)
- Sören Kammel (BOSCH Research Center - Palo Alto, US)
- Jana Kosecka (George Mason University - Fairfax, US) [dblp]
- Kevin Köser (ETH Zürich, CH)
- Bastian Leibe (RWTH Aachen, DE) [dblp]
- Volker Paelke (IG - Castelldefels (Barcelona), ES)
- Marc Pollefeys (ETH Zürich, CH) [dblp]
- Davide Scaramuzza (ETH Zürich, CH) [dblp]
- Konrad Schindler (ETH Zürich, CH) [dblp]
- Monika Sester (Leibniz Universität Hannover, DE) [dblp]
- Sudipta Sinha (Microsoft Corporation - Redmond, US)
- Noah Snavely (Cornell University, US) [dblp]
- Henrik Stewenius (Google Switzerland, CH)
- Christoph Stiller (KIT - Karlsruher Institut für Technologie, DE) [dblp]
- Aparna Taneja (ETH Zürich, CH)
- George Vosselman (ITC - Enschede, NL)
- Jonathan Warrell (Oxford Brookes University, GB)
- Ruigang Yang (University of Kentucky - Lexington, US) [dblp]
- ai / robotics
- computer graphics / computer vision
- data bases / information retrieval
- data structures / algorithms / complexity
- Dynamic maps
- spatial data infrastructure
- map matching
- autonomous navigation
- spatial cognition
- data fusion
- information retrieval
- intelligent vehicles
- 3D scene perception
- scene understanding
- 3D reconstruction.