Smartphone-based indoor localization systems are increasingly needed in various types of applications. This paper proposed a novel WiFi and Vision integrated fingerprint (Wi-Vi Fingerprint) for accurate and robust indoor localization. The method consists of two steps of fingerprint mapping and fingerprint localization. In the mapping step, the Wi-Vi fingerprints for all the sampling sites are computed by using EXIT signs as landmarks. In the localization step, a multi-scale localization strategy is proposed, which includes coarse localization from Weighted Access Points (WAP) based WiFi matching, Gaussian Weighted KNN(GW-KNN) based image-level localization from holistic visual features, and finally metric localization for refinement. The proposed method has been tested in an indoor office building of 12,000 m2 and a mega-mall of 7,200 m2 with different types of smartphones. Experimental results demonstrate that the proposed method can achieve 95% and 98% site recognition rates from image-level localization. The final localization errors after metric localization are less than a half meter on average.

Self-localization is a crucial task for intelligent vehicles. Existing localization methods usually require high-cost IMU (Inertial Measurement Unit) or expensive LiDAR sensors (e.g., Velodyne HDL-64E). In this paper, we propose a low-cost yet accurate localization solution by using a custom-level GPS receiver and a low-cost camera with the support of HD map. Unlike existing HD map-based methods, which usually requires unique landmarks within the sensed range, the proposed method utilizes common lane lines for vehicle localization by using Kalman filter to fuse the GPS, monocular vision, and HD map for more accurate vehicle localization. In the Kalman filter framework, the observations consist of two parts. One is the raw GPS coordinate. The other is the lateral distance between the vehicle and the lane, which is computed from the monocular camera. The HD map plays the role of providing reference position information and correlating the local lateral distance from the vision and the GPS coordinates so as to formulate a linear Kalman filter. In the prediction step, we propose using a data-driven motion model rather than a Kinematic model, which is more adaptive and flexible. The proposed method has been tested with both simulation data and real data collected in the field. The results demonstrate that the localization errors from the proposed method are less than half or even one-third of the original GPS positioning errors by using low cost sensors with HD map support. Experimental results also demonstrate that the integration of the proposed method into existing ones can greatly enhance the localization results.

3D LiDAR maps are playing a more and more important role in intelligent vehicle localization. In this paper, the proposed SeqPolar localization method consists of two parts. The first part is a novel node-based LiDAR map representation method, termed polarized LiDAR map (PLM), and the second is a map matching algorithm based on a second-order hidden Markov model (HMM2) for vehicle localization. Unlike existing 3D LiDAR map, which is usually constructed by accumulating 3D LiDAR clouds collected at different times, the proposed polarized LiDAR map is generated from a set of nodes. Each node consists of three elements: a polarized LiDAR image, scene features extracted from the polarized LiDAR image, and sensor pose. The polarized LiDAR image encodes 3D coordinates and reflectivities of the 3D LiDAR cloud using a multi-channel image format, a more concise, straightforward, and structured representation of the 3D LiDAR cloud. In the localization step, we propose an HMM2-based method to match a sequence of input polarized images within the map nodes and find the nearest node from the PLM. Afterward, the vehicle is readily localized from the matched map node with 3D registration. The proposed SeqPolar localization method has been validated with the actual field dataset and the public KITTI database. Experimental results demonstrate that the proposed HMM2-based matching method can achieve up to 98% accuracy to find the nearest node from PLM. Moreover, the SeqPolar localization method based on the pre-built PLM can achieve 30-centimeter localization accuracy in average on both test datasets.

Automatic Guided Vehicles (AGVs) are crucial for improving the operational efficiency of industrial scenes, however the self-localization of AGVs faces severe challenges in a dynamic environment due to a large number of observation noises. To address the above issue, this paper presents a trusted area information based self-localization method to achieve efficient and accurate localization in a highly dynamic scene, where the proportion of dynamic obstacles is close to 50%. First of all, the “coarse-noise filter-fine” self-localization framework is proposed to establish the foundation of accurate and efficient localization. Then, the morphology-based trusted area selection is adopted to efficaciously extract the trusted area, in which the probability of appearing dynamic objects is low. After that, based on dynamic angular resolution filtering considering the trusted area, a robust point cloud generation is designed to generate trusted area information. Trusted area information contains dense reliable point cloud in the statice area and spares observation noise in the dynamic area, thus improving the performance of observation models and scan matching. Finally, the self-developed AGV is adopted to verify the effectiveness of the proposed self-localization system in real and simulated dynamic scenes.

This paper proposes a novel folding pattern for flexible extrinsic calibration of a camera and a 2D laser rangefinder (LRF). The proposed pattern consists of multiple foldable facets. By folding and unfolding the facets, we can easily adjust the sizes and expanding angles of the pattern so that it can be observed by sensors with different view angles at different distances. Moreover, the folding pattern can have different numbers of facets such that it can flexibly generate different constraints to simplify the calibration process. The proposed pattern is thus very flexible and practical. Another important feature is that the proposed method does not require computing the metric information of the pattern, such as plane normal and distance, as required in existing methods. Therefore, it can further simplify the calibration process and enhance the calibration accuracy. In this method, we first derived a linear solution, and then refined the solution with the particle swarm optimization (PSO) method by both enforcing the rotation constraint and minimizing the sum of the square of the point-to-line distances. The proposed calibration pattern and method have been validated with both simulation and real data. Experimental results show that the method is very flexible and accurate for the extrinsic calibration of a camera and a 2D LRF.