Postdoctoral Position in Vehicle Perception LabEx MS2T

Post doc Advisors: Franck DAVOINE, CNRS researcher, and Thierry DENOEUX, Professor. Heudiasyc laboratory, UMR CNRS-UTC 7253 Tel. +33 (0)3 44 23 52 79, firstname.lastname@hds.utc.fr Context of the study: The candidate will be based at the Heudiasyc laboratory in Compiègne (40 minutes by train from central Paris). Heudiasyc is a joint laboratory with CNRS – the National Center for Scientific Research – and the Université de Technologie de Compiègne. It was rated A+ (the highest rate) by the French National Research evaluation Agency. Heudiasyc fosters interdisciplinary research on information science and technology including machine learning, uncertain reasoning, operations research, robotics and knowledge management. In 2011, Heudiasyc was awarded a project of excellence on «Control of Technological Systems of Systems» (Labex MS2T), funded by the "Investment for the future" national program. The laboratory has strong relations with the automotive industry and advanced international academic research teams on intelligent vehicles. The successful candidate will join the Challenge Team DAPAD of the Labex MS2T, which involves 10 researchers and several Master and PhD students, focusing on Cooperative perception to support autonomous driving. He/she is expected to take an active role in the project, focusing on real driving scenarios with a fleet of four instrumented vehicles. Post-doc description: Today, driver assistance systems have entered the market, with successes like adaptive cruise control and lane-keeping assistance. Despite these successes, much research and development efforts are still necessary to achieve autonomous driving with complex automated tasks such as, e.g., overtaking and lane changing, crossing intersections, etc.. The development of such systems relies on multi-sensory perception functions, communication, information processing, automatic and adaptive learning. Intelligent vehicles have variable capabilities in terms of self-localization, perception of the driving environment or even prediction of trajectories of other road users. They can take advantage of perceptual or intentional information exchanged with other road users (such as vehicles, pedestrians or bicycles) via wireless communication, to augment their field of view and situational awareness of the dynamic traffic scene. Furthermore, vehicles can exploit additional information provided by roadside units, by the infrastructure, or high-level context given by digital maps. In this context, “V2X” communication can be seen as a supplementary virtual sensor. In collaboration with the DAPAD team members, the post-doc is expected to contribute to the definition, development and evaluation of a cooperative perception system in which the ego-vehicle will (1) perceive the scene with its own sensors, and (2) combine its local perception with raw or preprocessed information received from other road users and from the infrastructure. The construction of an efficient global fusion module requires a good model of uncertainty in the ego-vehicle perception, but also in the broadcasted information (transmission latency and range, timestamps, data correlation, etc.). Besides a closed test track, Heudiasyc laboratory hosts a fleet of instrumented autonomous vehicles dedicated to experiments, developments and integrations of technologies for autonomous driving. Tests and evaluations will be carried out through different scenarios (such as crossroads, overtaking or roundabout) involving three to four vehicles, in close collaboration with DAPAD team members working in the field of machine learning, driving scene perception with Lidar, radar as well as cameras, V2X communication, uncertainty modeling and distributed information fusion in vehicular ad hoc networks. Start and Duration: As soon as possible from Sept. 2016, the position is for one year with possibility of an extension, pending funding. How much will you earn? Monthly net salary ~ 2050€ (potentially higher with some teaching activities). Affordable housing easy to find in Compiegne. Your profile: PhD degree (preferably in Computer Science or Applied Mathematics). Solid programming skills; the project involves programming in C, C++, Matlab, Python… Solid mathematics knowledge (especially linear algebra and statistics). Creative and highly motivated. Fluent in English, both written and spoken. Prior knowledge in the areas of computer vision or machine learning is a plus. Documents required to apply: Send to franck.davoine@hds.utc.fr First contact: Curriculum vitae; A statement of research experience and interests; Short motivation letter. In a second stage: At least two references or recommendation letters. Location: Université de Technologie de Compiègne (UTC) Laboratory Heudiasyc UMR CNRS-UTC 7253 Centre de recherche de Royallieu, Avenue de Landshut CS 60319 - 60203 Compiègne Cedex – France https://www.hds.utc.fr Some references: Rudy Bunel, Franck Davoine and Philippe Xu. Detection of Pedestrians at Far distance. In Proceedings of the IEEE International Conference on Robotics and Automation, Stockholm, Sweden, May 16-21, 2016. Philippe Xu, Franck Davoine, Jean-Baptiste Bordes, Huijing Zhao, Thierry Denoeux. Multimodal Information Fusion for Urban Scene Understanding. Machine Vision and Applications, Springer, 27(3), Pages 331-349, 2016. Philippe Xu, Franck Davoine, Hongbin Zha, Thierry Denoeux, Evidential calibration of binary SVM classifiers, International Journal of Approximate Reasoning, Elsevier, Volume 72, Pages 55-70, May 2016. Javier Radak, Bertrand Ducourthial, Veronique Cherfaoui, Stephane Bonnet. Detecting road events using distributed data fusion: experimental evaluation for the icy roads case, IEEE Intelligent Transportation Systems Transactions, 17, 1, 2016. Philippe Xu, Franck Davoine, and Thierry Denoeux, Evidential combination of pedestrian detectors, BMVC - British Machine Vision Conference, Nottingham, UK, Sept. 1-5, 2014. Nicole El Zoghby, Veronique Cherfaoui, Thierry Denoeux. Evidential Distributed Dynamic Map for Cooperative Perception in VANets. IEEE intelligent Vehicles Symposium 2014, Dearborn, Michigan, USA, 2014. Recent news about us: 1) In Journal of CNRS: https://news.cnrs.fr/articles/should-driverless-cars-also-be-borderless 2) We successfully participated to the Grand Cooperative Driving Challenge (GCDC’16) in The Netherlands, May 28-29, 2016. http://www.gcdc.net/en/event-en If you apply for this position please say you saw it on Computeroxy

Post doc Advisors:

Franck DAVOINE, CNRS researcher, and Thierry DENOEUX, Professor.
Heudiasyc laboratory, UMR CNRS-UTC 7253
Tel. +33 (0)3 44 23 52 79, firstname.lastname@hds.utc.fr

Context of the study:

The candidate will be based at the Heudiasyc laboratory in Compiègne (40 minutes by train from central Paris). Heudiasyc is a joint laboratory with CNRS – the National Center for Scientific Research – and the Université de Technologie de Compiègne. It was rated A+ (the highest rate) by the French National Research evaluation Agency. Heudiasyc fosters interdisciplinary research on information science and technology including machine learning, uncertain reasoning, operations research, robotics and knowledge management. In 2011, Heudiasyc was awarded a project of excellence on «Control of Technological Systems of Systems» (Labex MS2T), funded by the "Investment for the future" national program. The laboratory has strong relations with the automotive industry and advanced international academic research teams on intelligent vehicles.

The successful candidate will join the Challenge Team DAPAD of the Labex MS2T, which involves 10 researchers and several Master and PhD students, focusing on Cooperative perception to support autonomous driving. He/she is expected to take an active role in the project, focusing on real driving scenarios with a fleet of four instrumented vehicles.

Post-doc description:

Today, driver assistance systems have entered the market, with successes like adaptive cruise control and lane-keeping assistance. Despite these successes, much research and development efforts are still necessary to achieve autonomous driving with complex automated tasks such as, e.g., overtaking and lane changing, crossing intersections, etc.. The development of such systems relies on multi-sensory perception functions, communication, information processing, automatic and adaptive learning.

Intelligent vehicles have variable capabilities in terms of self-localization, perception of the driving environment or even prediction of trajectories of other road users. They can take advantage of perceptual or intentional information exchanged with other road users (such as vehicles, pedestrians or bicycles) via wireless communication, to augment their field of view and situational awareness of the dynamic traffic scene. Furthermore, vehicles can exploit additional information provided by roadside units, by the infrastructure, or high-level context given by digital maps. In this context, “V2X” communication can be seen as a supplementary virtual sensor.

In collaboration with the DAPAD team members, the post-doc is expected to contribute to the definition, development and evaluation of a cooperative perception system in which the ego-vehicle will (1) perceive the scene with its own sensors, and (2) combine its local perception with raw or preprocessed information received from other road users and from the infrastructure. The construction of an efficient global fusion module requires a good model of uncertainty in the ego-vehicle perception, but also in the broadcasted information (transmission latency and range, timestamps, data correlation, etc.).

Besides a closed test track, Heudiasyc laboratory hosts a fleet of instrumented autonomous vehicles dedicated to experiments, developments and integrations of technologies for autonomous driving. Tests and evaluations will be carried out through different scenarios (such as crossroads, overtaking or roundabout) involving three to four vehicles, in close collaboration with DAPAD team members working in the field of machine learning, driving scene perception with Lidar, radar as well as cameras, V2X communication, uncertainty modeling and distributed information fusion in vehicular ad hoc networks.

Start and Duration:

As soon as possible from Sept. 2016, the position is for one year with possibility of an extension, pending funding.

How much will you earn?

Monthly net salary ~ 2050€ (potentially higher with some teaching activities). Affordable housing easy to find in Compiegne.

Your profile:

PhD degree (preferably in Computer Science or Applied Mathematics). Solid programming skills; the project involves programming in C, C++, Matlab, Python… Solid mathematics knowledge (especially linear algebra and statistics). Creative and highly motivated. Fluent in English, both written and spoken. Prior knowledge in the areas of computer vision or machine learning is a plus.

Documents required to apply:

Send to franck.davoine@hds.utc.fr

First contact: Curriculum vitae;

A statement of research experience and interests;

Short motivation letter.

In a second stage: At least two references or recommendation letters.

Location:

Université de Technologie de Compiègne (UTC) Laboratory Heudiasyc UMR CNRS-UTC 7253
Centre de recherche de Royallieu, Avenue de Landshut
CS 60319 - 60203 Compiègne Cedex – France
https://www.hds.utc.fr

Some references:

Rudy Bunel, Franck Davoine and Philippe Xu. Detection of Pedestrians at Far distance. In Proceedings of the IEEE International Conference on Robotics and Automation, Stockholm, Sweden, May 16-21, 2016.

Philippe Xu, Franck Davoine, Jean-Baptiste Bordes, Huijing Zhao, Thierry Denoeux. Multimodal Information Fusion for Urban Scene Understanding. Machine Vision and Applications, Springer, 27(3), Pages 331-349, 2016.

Philippe Xu, Franck Davoine, Hongbin Zha, Thierry Denoeux, Evidential calibration of binary SVM classifiers, International Journal of Approximate Reasoning, Elsevier, Volume 72, Pages 55-70, May 2016.

Javier Radak, Bertrand Ducourthial, Veronique Cherfaoui, Stephane Bonnet. Detecting road events using distributed data fusion: experimental evaluation for the icy roads case, IEEE Intelligent Transportation Systems Transactions, 17, 1, 2016.

Philippe Xu, Franck Davoine, and Thierry Denoeux, Evidential combination of pedestrian detectors, BMVC - British Machine Vision Conference, Nottingham, UK, Sept. 1-5, 2014.

Nicole El Zoghby, Veronique Cherfaoui, Thierry Denoeux. Evidential Distributed Dynamic Map for Cooperative Perception in VANets. IEEE intelligent Vehicles Symposium 2014, Dearborn, Michigan, USA, 2014.

Recent news about us:

1) In Journal of CNRS: https://news.cnrs.fr/articles/should-driverless-cars-also-be-borderless

2) We successfully participated to the Grand Cooperative Driving Challenge (GCDC’16) in The Netherlands, May 28-29, 2016. http://www.gcdc.net/en/event-en

If you apply for this position please say you saw it on Computeroxy