Search

The use of bicycles is regaining popularity, especially in city centers where they can be used as quickly as cars and reduce carbon footprint. However, in these dense urban environments, navigation methods based on GNSS technologies do not provide sufficient accuracy for cyclist navigation and safety. To mitigate the challenges of indoor-like surroundings, a new positioning algorithm: BIKES (Bicycle Itinerancy Kalman filter with Embedded Sensors) was developed. This extended Kalman filter processes deeply degraded GNSS data to update velocity and position estimates with differential computation approaches working even in degraded environments. GNSS signals are combined with inertial and magnetic data to continuously estimate the trajectory when GNSS is unavailable. BIKES' performance was tested in real-life conditions on a 3 km long path in the city center downtown Nantes and compared to Google Fused Location Provider estimates. A mean positioning error below 1 m with a 0.5 m standard deviation is achieved. These results are 4 times better than the Google solution. This algorithm allows also distinguishing if the cyclist is riding on a bike path, the sidewalk, or the road, which is critical for guidance systems.

Pedestrian navigation with handheld sensors is still particularly complex. Pedestrian Dead Reckoning method is generally used, but the estimation of the walking direction remains problematic because the device's pointing direction does not always correspond to the walking direction. To overcome this difficulty, it is possible to use gait modeling based approaches. But, these methods suffer from sporadic erroneous estimates and their accumulation over time. The HEAD (smootH Estimation of wAlking Direction) filter uses WAISS and MAGYQ angular estimates as observations to correct the walking direction and to obtain more robust and smooth results. TDCP updates are applied to constrain the walking direction estimation error while pseudo-ranges directly update the position. HEAD is tested by 5 subjects over 21 indoor/outdoor acquisitions (between 720 m and 1.3 km). A 54% improvement is achieved thanks to the fusion in texting mode. The median obtained angular error is 5.5 degrees in texting mode and 12 degrees in pocket mode.

32nd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2019), MIAMI, ETATS-UNIS, 16-/09/2019 - 20/09/2019; Pedestrian location in urban or indoor environments is particularly complex. Indeed, GNSS technology generally used for localization is no longer sufficient in these challenging environments. However, the presence of many sensors in consumer equipment like smartphones allows the implementation of different methods. PDR (Pedestrian Dead Reckoning) is a position estimation method using inertial and magnetic sensor data. It is based on the determination of two elements: the step length and the walking direction. This direction is difficult to estimate for handheld sensors because the orientation of the sensor is not always aligned with the walking direction. Methods based on the study of horizontal hand accelerations can overcome this difficulty, but performance on real scenarios is not sufficient. This article proposes a new method for estimating the walking direction and position based on an extended Kalman filter. For this purpose, the angular estimates from the WAISS and MAGYQ algorithms are merged to update the estimate of the walking direction. Phase measurements are used with TDCP updates to correct the velocity and correct the walking direction. 6 experiments carried out with three subjects over distances between 650 and 1300m in texting mode, in real and challenging conditions are conducted. The mean angular error obtained is 4.6° and the mean position error is 0.5% of the travelled distance

2018 International Conference on Indoor Positioning and Indoor Navigation, NANTES, FRANCE, 24-/09/2018 - 27/09/2018; Inertial and magnetic sensors based PDR approaches are particularly interesting for pedestrian location since they don't require any specific infrastructure. Estimating the walking direction, which is essential for PDR strategy, remains difficult with handheld sensors. WAISS is a new method that integrates hand movement and is independent of the misalignment between the walking direction and the pointing direction that estimates the walking direction. It uses statistical models of the hand accelerations in the horizontal plane. The paper studies how to create the best possible models. Among the features under study are the number of strides used to learn the models, different acquisition contexts and walking directions. Finally, the complexity of models needed for a given person is discussed. 100 strides over curved and straight line walks combined with a bi-modal Gaussian Mixture Model gives the best walking direction estimate with a 15° mean error over a 325 m indoor/outdoor walk performed by four subjects.

Depuis quelques années des compétitions internationales destinées à comparer les technologies de localisation à l'intérieur des bâtiments sont organisées. Face à la diversification de ces technologies, elles permettent de fixer une cadre unique d'évaluation des performances de localisation en temps réel ou différé. Un levé topographique d'envergure qui combine mesures au théodolite, par GNSS différentiel et scanner 3D a permis de cartographier à 10 cm près les 180 cibles réparties dans le centre commercial Atlantis à Nantes. Ces cibles définissent les parcours sur lesquels les compétiteurs du congrès international IPIN s'affronteront le 22 septembre. Ce projet a été réalisé par quatre étudiants de l'ESGT sous la direction du laboratoire GEOLOC de l'IFSTTAR et avec le soutien de la société Viametris.