Companion robots, such as the humanoid robot NAO manufactured by Aldebaran Robotics, should be able to move on non-planar surfaces, such as going up- or downstairs. This implies that the robot is able to perceive the floor and to extract piecewise planar surfaces, such as the stairs, independently of the floor type. This can be done using a 3D sensor, such as the Microsoft Kinect or a stereoscopic camera pair, and an algorithm that is able to take as input a 3D image and to produce as output a segmentation of this image in terms of planar patches. For this purpose we plan to use a regression method recently developed in the Perception team, namely GLLiM. GLLiM stands for “Gaussian Locally-Linear Mapping’’. Currently we have a Matlab implementation of this method. The work of the student will consist of fully understanding the method and of adapting the existing Matlab code to the data and to the problem to be solved. A more efficient implementation (in C) could then be envisaged using the robot programming environment available in the team.
The project is suitable for a first year master student with very good background in applied mathematics as well as very good programming skills in Matlab and C/C++.
The project may start anytime after 1 February 2015 for a period of three and up to six months.
Information for applicants: Please send your complete CV, university grades, and the names and emails of two recommending persons to firstname.lastname@example.org. Students enrolled in a French university will receive a monthly net salary of 430€. Students enrolled in another university will receive a monthly net salary of 1100€. Please note that some restrictions apply to non-French students and their admission is conditioned by an approval from the French Ministry of Defense.