Denis Pomorski

GPS/IMU data fusion using multisensor Kalman filtering: introduction of contextual aspects

The aim of this article is to develop a GPS/IMU multisensor fusion algorithm, taking context into consideration. Contextual variables are introduced to define fuzzy validity domains of each sensor. The algorithm increases the reliability of the position information. A simulation of this algorithm is then made by fusing GPS and IMU data coming from real tests on a land vehicle. Bad data delivered by GPS sensor are detected and rejected using contextual information thus increasing reliability. Moreover, because of a lack of credibility of GPS signal in some cases and because of the drift of the INS, GPS/INS association is not satisfactory at the moment. In order to avoid this problem, the authors propose to feed the fusion process based on a multisensor Kalman filter directly with the acceleration provided by the IMU. Moreover, the filter developed here gives the possibility to easily add other sensors in order to achieve performances required.

Towards symbolization using data-driven extraction of local trends for ICU monitoring

We propose a methodology for the extraction of local trends from a stream of data. It has been designed to suit the needs of interpretation-oriented visualization and symbolization from ICU monitoring data. After giving implementation details for efficient computation of local trends, we propose the use of a characteristic analysis span for each variable. This characteristic span is obtained from a set of criteria that we compare and evaluate in regard of analysis of ICU monitoring data gathered within the Aiddaig project. The processing results in a rich visual representation and a framework for the local symbolization of the data stream based on its dynamics.

Virtual 3D City Model for Navigation in Urban Areas

In this paper, we propose to study the integration of a new source of a priori information, which is the virtual 3D city model. We study this integration for two tasks: vehicles geo-localization and obstacles detection. A virtual 3D city model is a realistic representation of the evolution environment of a vehicle. It is a database of geographical and textured 3D data. We describe an ego-localization method that combines measurements of a GPS (Global Positioning System) receiver, odometers, a gyrometer, a video camera and a virtual 3D city model. GPS is often consider as the main sensor for localization of vehicles. But, in urban areas, GPS is not precise or even can be unavailable. So, GPS data are fused with odometers and gyrometer measurements using an Unscented Kalman Filter (UKF). However, during long GPS unavailability, localization with only odometers and gyrometer drift. Thus, we propose a new observation of the location of the vehicle. This observation is based on the matching between the current image acquired by an on-board camera and the virtual 3D city model of the environment. We also propose an obstacle detection method based on the comparison between the image acquired by the on-board camera and the image extracted from the 3D model. The following principle is used: the image acquired by the on-board camera contains the possible dynamic obstacles whereas they are absent from the 3D model. The two proposed concepts are tested on real data.

Intelligent Geolocalization in Urban Areas Using Global Positioning Systems, Three-Dimensional Geographic Information Systems, and Vision

This article tackles the problem of a vehicles geolocalization in urban areas. For this purpose, Global Positioning System (GPS) receiver is the main sensor. However, the use of GPS alone is not sufficient in many urban environments. GPS has to be helped with dead-reckoned sensors, map data, and cameras. A novel observation of the absolute pose of the vehicle is proposed to back up GPS and the drift of dead-reckoned sensors. This approach uses a new source of information that is a geographical 3-dimensional (3D) model of the environment in which the vehicle navigates. This virtual 3D city model is managed in real time by a 3D geographical information system (3D GIS). This poses observation is constructed by matching the virtual image provided by the 3D GIS and the real image acquired by an onboard camera. An extended Kalman filter combines the sensors measurements to produce an estimation of the vehicles pose. Experimental results using data from an odometer, a gyroscope, a GPS receiver, a camera, and an accurate geographical 3D model of the environment illustrate the developed approach.

Inductive learning of decision trees: application to fault isolation of an induction motor

This work deals with fault detection and isolation (FDI) of an induction motor. Its supervision cannot be performed on the sole knowledge of analytical redundancy relations : a normal functioning state of the motor and a speed-sensor failure state cannot be distinguished from a behavioral analytical model. A solution is proposed using two inductive learning techniques based on decision tree formalism: C4.5 which is a milestone in top–down induction of decision trees, and BUST which is a solution for the functional separability problem of decision trees.

Localisation in urban environment using GPS and INS aided by monocular vision system and 3D geographical model

In this paper, a geo-localisation method was proposed, using GPS, INS, monovision camera and a new geo-information source, which is the 3D cartographical model. A 3D-GIS (Geographical Information System) has been developed to manipulate and navigate in a precise 3D cartographical model database. To have a continuous pose estimation, an EKF has been implemented to fuse GPS and INS. To integrate the 3D cartographical model information, a 3D cartographical observation has been constructed using 2D/3D images matching between the real image captured by the embedded camera and the virtual images provided by the 3D-GIS. A real data acquisition platform has been developed to test and validate the proposed method.

Harris, SIFT and SURF features comparison for vehicle localization based on virtual 3D model and camera

This paper proposed a new vehicle geo-localization method in urban environment integrating a new source of information that is a virtual 3D city model. This 3D model provides a realistic representation of the navigation environment of the vehicle. To optimize the performance of vehicle geo-localization system, several sources of information are integrated for their complementarity and redundancy: a GPS receiver, proprioceptive sensors (odometers and gyrometer), a video camera and a virtual 3D city model. The pose estimation algorithm used to fuse the different sensors data is an IMM-UKF (Interacting Multiple Model - Unscented Kalman Filter). The proprioceptive sensors allow to continuously estimating the dead-reckoning position and orientation of the vehicle. This dead-reckoning estimation of the pose is corrected by GPS measurements. Moreover, a 3D model/camera based observation of the vehicle pose is constructed to compensate the drift of the dead-reckoning localization when GPS measurements are unavailable for a long time. This pose observation is based on the matching between the virtual image extracted from the 3D city model and the real image acquired by the camera. The observation construction is composed of two major parts. The first part consists in detecting and matching the feature points of the real and virtual images. Three features are compared: Harris corner, SIFT (Scale Invariant Feature Transform) and SURF (Speed Up Robust Features). The second part is the pose computation using POSIT algorithm and the previously matched features set. The developed approach has been tested on a real sequence and the obtained results proved the feasibility and robustness of the approach.

Vehicle geo-localization based on IMM-UKF data fusion using a GPS receiver, a video camera and a 3D city model

Vehicle geo-localization in urban areas remains to be challenging problems. For this purpose, Global Positioning System (GPS) receiver is usually the main sensor. But, the use of GPS alone is not sufficient in many urban environments due to wave multi-path. In order to provide accurate and robust localization, GPS has so to be helped with other sensors like dead-reckoned sensors, map data, cameras or LIDAR. In this paper, a new observation of the absolute pose of the vehicle is proposed to back up GPS measurements. The proposed approach exploits a virtual 3D model managed by a 3D geographical information system (3D GIS) and a video camera. The concept is to register the acquired image to the 3D model that is geo-localized. For that, two images have to be matched: the real image and the virtual image. The real image is acquired by the on board camera and provides the real view of the scene viewed by the vehicle. The virtual image is provided by the 3D GIS. The developed method is composed of three parts. The first part consists in detecting and matching the feature points of the real image and of the virtual image. Two methods: SIFT (Scale Invariant Feature Transform) and Harris corner detector are compared. The second part concerns the position computation using POSIT algorithm and the previously matched features set. The third part concerns the data fusion using IMM-UKF (Interacting Multiple Model-Unscented Kalman Filter). The proposed approach has been tested on a real sequence and the obtained results proved the feasibility and robustness of the approach.

Outdoor Obstacle Detection and Localisation with Monovision and 3D Geographical Database

In this paper, an obstacle detection approach for downtown environments is developed. This approach exploits a 3D geographical database managed by a 3D-GIS and a monovision-based system. The pose estimated by a LRK GPS is used to geo-localise the vehicle. After coordinates system conversion, the vehicle is localised in the 3D geographical database. An image processing module is developed to match synchronized images provided by 3D GIS and an on-board camera. Several kinds of obstacles are then detected and tracked by comparison between real images and virtual images. Finally, the distance between the camera and the obstacles is computed, as well as the geo-position of the detected obstacles. Experimental results with real data are presented in the final section.

Fault tolerant collaborative localization for multi-robot system

Multi-robot system is used in some unreachable or dangerous area in order to replace the human operators. In such environments the integrity of localization should be assured by adding a sensor fault diagnosis step. In this paper, we present a method able, in addition of localizing a group of robots, to detect and exclude the faulty sensors from the team. The estimator is the informational form of the Kalman Filter (KF) namely Information Filter (IF). The developed residual test is based on the divergence between the predicted and the corrected estimation of the IF, calculated in term of the Kullback-Leibler divergence (KLD). The main contributions of this paper: - developing a method able simultaneously to localize a group of robots and to detect the faulty sensors - using the IF and the KLD as a residual test - Application of the proposed framework to a real environment with real robots.