Stéphane Côté

Indoor Pedestrian Navigation Using Foot-Mounted IMU and Portable Ultrasound Range Sensors

Many solutions have been proposed for indoor pedestrian navigation. Some rely on pre-installed sensor networks, which offer good accuracy but are limited to areas that have been prepared for that purpose, thus requiring an expensive and possibly time-consuming process. Such methods are therefore inappropriate for navigation in emergency situations since the power supply may be disturbed. Other types of solutions track the user without requiring a prepared environment. However, they may have low accuracy. Offline tracking has been proposed to increase accuracy, however this prevents users from knowing their position in real time. This paper describes a real time indoor navigation system that does not require prepared building environments and provides tracking accuracy superior to previously described tracking methods. The system uses a combination of four techniques: foot-mounted IMU (Inertial Motion Unit), ultrasonic ranging, particle filtering and model-based navigation. The very purpose of the project is to combine these four well-known techniques in a novel way to provide better indoor tracking results for pedestrians.

The Use of the Hopfield Neural Network to Measure Sea-Surface Velocities From Satellite Images

The knowledge of ocean surface circulation is of major importance for many applications, including the understanding of global climate, resources exploitation, and containment of chemical spills. In this letter, sea-surface feature tracking based on the Hopfield neural network (NN) is described. The method is based on the minimization of an energy function that represents the feature tracking problem. A Hopfield NN is used to merge cross-correlation information with prior knowledge of sea-surface flows and image contextual information. It has been tested on real satellite images. A set of five Advanced Very High Resolution Radiometer thermal images of the coastal zone of California, along with a data set of coincident surface drifters positions, was used to test the method. Results of the new analysis are compared with in situ data and previous results using other techniques. The method can be used on various kinds of images for tracking and also find other applications in image registration and pattern recognition.

A live Augmented Reality Tool for Facilitating Interpretation of 2D Construction Drawings

Construction consists of a complex set of tasks in the 3D world based on instructions encoded in 2D drawings. Although the process is facilitated by the availability of 3D models displayed on digital tablets on construction sites, it is not always clear what is the exact 3D location of specific elements in 2D drawings. In this preliminary study, we propose a method based on a computer tablet and a head mounted augmentation system that enables the user to display a 3D element by clicking on its 2D representation on a construction drawing. Early results show that the method has potential, but highlights perception issues. We proposed and tested solutions to alleviate those issues.

[Poster] Augmentation of live excavation work for subsurface utilities engineering

The virtual excavation is a well-known augmentation technique that was proposed for city road environments. It can be used for planning excavation work by augmenting the road surface with a virtual excavation revealing subsurface utility pipes. In this paper, we proposed an extension of the virtual excavation technique for live augmentation of excavation work sites. Our miniaturized setup, consisting of a sandbox and a Kinect device, was used to simulate dynamic terrain topography capture. We hypothesized that the virtual excavation could be used live on the ground being excavated, which could facilitate the excavator operators work. Our results show that the technique can indeed be adapted to dynamic terrain topography, but turns out to occlude terrain in a potentially hazardous way. Potential solutions include the use of virtual paint markings instead of a virtual excavation.

Third person perspective augmented reality for high accuracy applications

We proposed a 6 degrees of freedom augmentation system aimed at meeting the high accuracy requirements of engineering tasks. A stationary panoramic video camera captures a stream that is augmented by a portable computer. A handheld tablet device located in the same area broadcasts its instantaneous orientation, and receives the augmented view in the corresponding orientation, in real time. The panoramic camera can also be moved to other locations and simultaneously tracked by the system, providing 6 degrees of freedom augmentation. This gives the user a third person perspective augmentation, which is very precise and potentially more accurate than handheld augmentation.

A building-wide indoor tracking system for augmented reality

Buildings require regular maintenance, and augmented reality (AR) could advantageously be used to facilitate the process. However, such AR systems would require accurate tracking to meet the needs of engineers, and work accurately in entire buildings. Popular tracking systems based on visual features cannot easily be applied in such situations, because of the limited number of visual features indoor, and of the high degree of similarity between rooms. In this project, we propose a hybrid system combining low accuracy radio-based tracking, and high accuracy tracking using depth images. Results show tracking accuracy that would be compatible with AR applications.

Accurate OnSite Georeferenced Subsurface Utility Model Visualisation

Subsurface utilities, essential to our modern life, must be regularly maintained to ensure continued operation. Subsurface utility engineering would greatly benefit from augmented reality, which would provide workers with direct up to date information about the underground pipes. Unfortunately, while accuracy is very important for engineers, outdoor augmented reality cannot yet guarantee a constant and known level of accuracy. In this paper, we propose a system based on the use of a robotic total station that provides the users georeferenced position in real time. A tablet displays a 3D rendering of the subsurface utility network at the users location, enabling him to explore the network without having to carry paper plans and make measurements. The system highlights several potential advantages over classical measuring methods, as well as with augmented reality.