Vision and machine learning for robot manipulation

The Perception and Activity Understanding group (Jean-Marc Odobez, http://www.idiap.ch/~odobez/), along with the Robot Learning and Interaction group (Sylvain Calinon, http://calinon.ch) at Idiap (www.idiap.ch), seeks one motivated postdoctoral fellow to work on interaction strategies and deep learning for object manipulations in cluttered object scenes.

The position is part of a European CHIST-ERA project involving the University of Lincoln (UK), INRIA (France), TU Wien (Austria), and IIT (Italy). The project aims at benchmarking object recognition, manipulation and human-robot interaction for sorting a complex unstructured heap of unknown and irregular objects. The Idiap research institute will focus on pushing or pulling tasks and in combining different interaction strategies (self-learning, learning from demonstration, learning by correction) including the use of deep learning approaches to model multimodal action-perception interactions (proprioception vs visual modalities).

The ideal postdoctoral candidate should hold a MS degree in computer science, engineering, physics or applied mathematics. S/he should have an excellent background in statistics, linear algebra, signal processing, programming, machine learning. The successful applicant will have good analytical skills, written and oral communication skills, and the ability to work and collaborate in a project with international teams. The recruited person will complement and work with a number of other teams members of the two research groups involved in related topics (e.g. in the SNSF funded ROSALIS project).

Research during the Phd or/and prior substantial experience in at least two of the three following domains is imperative, along with the willingness to learn the third one:
- robotics (related to gesture and/or manipulation),
- machine learning,
- computer vision.

More specific expertise, experience or technical skills sought include:
- robot manipulation (gesture learning and synthesis, grasping)
- visual servoing
- active perception
- deep learning
- visual processing
- software: python and/or C++, ROS

The position is for one year, with possibilities of being renewed until the end of the project. The Annual gross salary is 80,000 CHF.

The project will start in March 2019, but the position can start earlier or later, depending on the profile.

Interested candidates should submit a cover letter, a detailed CV, and the names of three references (or recommendation letters) through the Idiap online recruitment system:

Vision and machine learning for robot manipulation

Interviews will start upon reception of applications until the position is filled.