Computer Vision Augmented Geospatial Localization

US Patent, 2018

According to the embodiments provided herein, a trajectory determination device for geo-localization can include one or more relative position sensors, one or more processors, and memory. The one or more processors can execute machine readable instructions to receive the relative position signals from the one or more relative position sensors. The relative position signals can be transformed into a sequence of relative trajectories. Each of the relative trajectories can include a distance and directional information indicative of a change in orientation of the trajectory determination device. A progressive topology can be created based upon the sequence of relative trajectories; this progressive topology can be compared to map data. A geolocation of the trajectory determination device can be determined.

International Journal of Control, Automation and Systems, 2011

GPS/INS integrated systems do not guarantee robustness and accuracy of localization, because GPS has vulnerability to external disturbances. However, the overall performance and reliability of the system can be significantly improved by fusing multiple sensors with a different operating principle. In outdoor environments where GPS may be blocked, there are many features compared to the open space and these features can provide much information for UGV localization. Thus, this paper proposes an improved localization algorithm based on the hierarchical federation of three measurement layers, i.e., GPS, INS, and visual localization, to overcome the shortcomings of GPS/INS integrated systems. The proposed algorithm automatically switches the operation modes according to GPS status and a network of a ground-based reference station. A vocabulary tree with SURF is used in the visual localization method. In the data fusion of visual localization and INS, an asynchronous and timedelayed data fusion algorithm is presented because visual localization is always time-delayed compared with INS. By using DGPS to obtain the reference position under the dynamic conditions of the reference station, the restrictions of the conventional DGPS are overcome and all UGVs within WiBro communication range of the reference station can accurately estimate the position with a common GPS. The experiment results with a predefined path demonstrate enhancement of the robustness and accuracy of localization in outdoor environments.

Annals of ISPRS, 2016

The primary method for geo-localization is based on GPS which has issues of localization accuracy, power consumption, and unavailability. This paper proposes a novel approach to geo-localization in a GPS-denied environment for a mobile platform. Our approach has two principal components: public domain transport network data available in GIS databases or OpenStreetMap; and a trajectory of a mobile platform. This trajectory is estimated using visual odometry and 3D view geometry. The transport map information is abstracted as a graph data structure, where various types of roads are modeled as graph edges and typically intersections are modeled as graph nodes. A search for the trajectory in real time in the graph yields the geo-location of the mobile platform. Our approach uses a simple visual sensor and it has a low memory and computational footprint. In this paper, we demonstrate our method for trajectory estimation and provide examples of geolocalization using public-domain map data. With the rapid proliferation of visual sensors as part of automated driving technology and continuous growth in public domain map data, our approach has the potential to completely augment, or even supplant, GPS based navigation since it functions in all environments.

2006 IEEE Intelligent Vehicles Symposium, 2006

This paper addresses 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, we propose a strategy to manage this information in a GIS (Geographical Information System). We illustrate our developments using visual landmarks made of key images and 3D points that are regrouped in local maps 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.

2003

recognition phase, described in Section III, is accomplished by a voting scheme. Section IV describes briefly a novel robust estimation technique used to identify correct matches and estimate motions between the query view and reference views. Two reference views are then selected for the final GPS location triangulation. Individual steps of the approach are described in the following sections.

Medium precision positioning (below 50 cm) can be a valuable support in many surveying techniques as, e.g., Ground Penetrating Radar (GPR). GPR is an active instrument that uses the radar technology to penetrate the ground up to a few meters. Analysing the echo collected by the detector, a picture of the different kind of materials or structures underground can be produced. Among other applications, the detection of pipes in urban streets is of great importance, since in most cases there are no accurate maps of the underground network of cables and pipes laid by public utilities. This makes it difficult planning new installations and maintenance of the existing; when digging, damages and interference to the service often occur. To draw a plan of the underground status quo, a PR installed on a trailer is driven along the roads. At each pass, a strip about 2m large can be surveyed; operating speed can be as high as 15-20 km/h, but the normal daily productivity is about 3 km. In order to provide a useful map to users, the georeferencing accuracy of the GPR data should be at maximum around 20 cm. This accuracy target might be achieved today by several positioning system that rely on GPS or GNSS as a key (or as the sole) system component. Indeed, real-time as well as post processed kinematic positioning within a network of permanent GPS stations (NRTK or PPK) can deliver this accuracy. Nevertheless, the specific conditions of GPR urban surveys, make it impossible to rely on GPS alone for positioning. In front of large blocks, in narrow streets, or in tree-lined avenues it is well known that GPS positioning (even should the code solution be accurate enough) is not sufficient to guarantee uninterrupted coverage, because the number of visible satellites can often be less then 4. Besides the time to fix after a complete GPS outage can vary from 20 s to more than 200 s (Petovello et al., 2003), extending the duration of the GPS outages and sometimes forcing to stop the vehicle during the survey to recover the GPS solution. Usually, to overcome this limit, fusion of data collected from different sensors is operated. The most used integrated system combines GPS and inertial navigation system (INS) data performing the so called GPS-aided inertial navigation, adding to the accuracy of positioning the resistance to short term GPS outages (Zhang et al., 2005). By mean of the tightly coupled solution, the most recent systems allow a positioning precision of more than 1m after almost 1 min of complete absence of GPS signal. Besides, inertial aided navigation is currently still rather expensive. Other cheap instruments used to aid the GPS vehicle positioning are odometers, but their precision is not sufficient for our purposes.So our idea is to introduce photogrammetry to determine the positions of the vehicle, aided by GPS when available. In urban environment, the coordinates of some points located at every side of the surveying area are known with sufficient precision, thanks to the urban maps: this information can be used to georeference the photogrammetric strip.

2006

In this paper we present a prototype system for image based localization in urban environments. Given a database of views of city street scenes tagged by GPS locations, the system computes the GPS location of a novel query view. We first use a wide-baseline matching technique based on SIFT features to select the closest views in the database. Often due to a large change of viewpoint and presence of repetitive structures, a large percentage of matches (> 50%) are not correct correspondences. The subsequent motion estimation between the query view and the reference view, is then handled by a novel and efficient robust estimation technique capable of dealing with large percentage of outliers. This stage is also accompanied by a model selection step among the fundamental matrix and the homography. Once the motion between the closest reference views is estimated, the location of the query view is then obtained by triangulation of translation directions. Approximate solutions for cases when triangulation cannot be obtained reliably are also described. The presented system is tested on the dataset used in ICCV 2005 Computer Vision Contest and is shown to have higher accuracy than previous reported results.

2011 IEEE Intelligent Vehicles Symposium (IV), 2011

A global positioning method based on a precise 3-D drivable area map and on GPS pseudorange measurements is presented. Map and GPS measurements are represented by geometric constraints, thus turning the localization problem into a constraint satisfaction problem whose solution is the confidence domain of position. Interval analysis is employed to solve the problem by using contractions and bisections of a prior position box. If more than 3 satellites are visible, the method is robust to wrong pseudorange measurements. The system is also able to compute multiple position hypotheses in the case of ambiguities. An experimental validation using real GPS pseudorange measurements and a precise 3-D map is reported to illustrate the performance of the method with real data in an urban area, with reduced satellite visibility. Confidence domains are consistent with the truth during the whole 1 km experiment, and a 6.5 m 95% accuracy is achieved with at least two satellites in view.

Localization with respect to a reference model is a key feature for mobile robots. Urban environment offers numerous landmarks that can be used for the localization process. This paper deals with the use of an environment model stored in a Geographic Information System, to drive a vision system i.e. highlights what to look for ? and where to look for ? This task is achieved by propagating uncertainties along the image acquisition system.

2016

Given a picture taken somewhere in the world, automatic geo-localization of such an image is an extremely useful task especially for historical and forensic sciences, documentation purposes, organization of the world’s photographs and in-telligence applications. While tremendous progress has been made over the last years in visual location recognition within a single city, localization in natural environ-ments is much more difficult, since vegetation, illumination, seasonal changes make appearance-only approaches impractical. In this work, we target mountainous terrain and use digital elevation models to extract representations for fast visual database lookup. We propose an automated approach for very large scale visual localization that can efficiently exploit visual information (contours) and geometric constraints (consistent orientation) at the same time. We validate the system at the scale of Switzerland (40000km2) using over 1000 landscape query images with ground truth GPS pos...