UMR CNRS 7253

Site Tools


en:video

This is an old revision of the document!


This page contains several videos

Drivable space characterization

<html> <iframe width=“640” height=“360” src=“http://www.youtube.com/embed/d4Bsrbnawxw” frameborder=“0” allowfullscreen></iframe> </html>

The characterization in real-time of the drivable space in front of the vehicle is a key issue for safe autonomous navigation or driving assistance. This paper presents a method that uses a lidar (a multilayer laser scanner) integrated in the front bumper of an automotive vehicle. A grid processing is first applied to detect and localize objects in the immediate environment after having compensated of the movement of the vehicle. Accurate map information is then introduce in the perception scheme to refine the characterization of the drivable space. The paper details the different processing stages necessary to implement this method and presents the design of the system that has been prototyped on board an experimental vehicle. We report real experiments carried out in challenging urban environments to illustrate the performance of this approach which has been evaluated thanks to a precise retro-projection of the estimated drivable space in a wide-angle scene camera.

Use of map data in GNSS computation fix

<html> <object width=“425” height=“350”> <param name=“movie” value=“http://www.youtube.com/v/ubjhBwpy8pA”> </param> <embed src=“http://www.youtube.com/v/ubjhBwpy8pA” type=“application/x-shockwave-flash” width=“425” height=“350”> </embed> </object> </html>

In this video, we present a data fusion strategy relying principally on DR under the hypothesis that the DR performances are very good. When the localizer is tracking the pose, the map and the GPS observations are considered as doubtful. They are involved in the localisation process when they are consistent with the DR pose. This is carried out using integrity Chi-Square checks on innovations signals. When the vehicle approaches an ambiguity zone (detected thanks to the connectivity information of the map), the map information is not taken into account. If the map outage is too long, GPS data is also forgotten since it is not possible to detect signal propagation troubles. When leaving the ambiguity zone, the complete tracking starts again and checks the integrity of all the available information. In practice, we have developed a tightly-coupled Kalman filter which switches between the different observations modes.

Visual memory management

<html> <object width=“425” height=“344”><param name=“movie” value=“http://www.youtube.com/v/55gbgNaZQlk&hl=en&fs=1”></param><param name=“allowFullScreen” value=“true”></param><embed src=“http://www.youtube.com/v/55gbgNaZQlk&hl=en&fs=1” type=“application/x-shockwave-flash” allowfullscreen=“true” width=“425” height=“344”></embed></object> </html>

This video Shows the problem of localizing a vehicle in urban environment by using natural information provided by exteroceptive sensors. For this purpose, sensors need to detect landmarks which have been characterized in a previous passage. As the amount of data can be significantly large, a strategy has been used to manage this information in a GIS (Geographical Information System). The developments have been illustrated using visual landmarks made of key images and 3D points that correspond to the roads of a GIS layer thanks to the use of GPS data and proprioceptive sensors. Real experiments are reported to illustrate the performance of this approach which is robust to GPS outages due to poor satellite visibility in urban areas.


User Tools