Research on a Novel Indoor Localization Method for Mobile Robot

This paper aims to design and implement an indoor localization method for mobile robots based on trilateration technique, Color Code signatures (CCs) as artificial passive beacons, and a depth camera. The artificial passive beacons are designed as three disks positioned one on top of the other’s with different sizes and colors (primary colors only, RGB). The three disks share the same center. The designed beacons are placed on the ceiling to be visible from most locations within the room. The robot's model, controller, and localization method are implemented and evaluated inside the CoppeliaSim environment. The simulation results show that the CCs detecting algorithm enables the robot to discover the designed beacons, achieve an accurate localization and reach the target

In this paper, we propose a novel method for mobile robot localization and navigation based on multispectral visual odometry (MVO). The proposed approach consists in combining visible and infrared images to localize the mobile robot under different conditions (day, night, indoor and outdoor). The depth image acquired by the Kinect sensor is very sensitive for IR luminosity, which makes it not very useful for outdoor localization. So, we propose an efficient solution for the aforementioned Kinect limitation based on three navigation modes: indoor localization based on RGB/depth images, night localization based on depth/IR images and outdoor localization using multispectral stereovision RGB/IR. For automatic selection of the appropriate navigation modes, we proposed a fuzzy logic controller based on images’ energies. To overcome the limitation of the multimodal visual navigation (MMVN) especially during navigation mode switching, a smooth variable structure filter (SVSF) is implemented to fuse the MVO pose with the wheel odometry (WO) pose based on the variable structure theory. The proposed approaches are validated with success experimentally for trajectory tracking using the mobile robot (Pioneer P3-AT).

Localization is one of many issues in mobile robot study. Localization is an essential ability for mobile robot to determine its location, so that it can plan a movement and go to a desired location. The mutual method for mobile robot localization is using a particle filter. High computation needs in particle filter is one of problem in particle filtering to get accurate location. The paper proposes a low computational mobile robot localization using a particle filter. It uses two methods: local localization using a dead reckoning method and global localization using a landmark-based vision sensor. Simulation results show that the proposed method provides a good estimation on the mobile robot position and orientation.

Mobile robot localization based on omnicamera

Owing to the presence of multiple obstacles or changes in the positions of surroundings objects, the localization of a mobile robot using only natural landmarks is difficult in indoor environments. Consequently, in these environments, the use of artificial landmarks to achieve robust localization is necessary. To this end, we propose a global localization method for mobile robots, using reflectors as indistinguishable artificial landmarks, not the manmade signs have a certain pattern. In previous studies, it has been shown that localization is possible using information about the distance between indistinguishable artificial landmarks. We applied this principle in a real-world situation with many obstacles or in a dynamic environment and confirmed that robust, high-accuracy global localization is possible.

This paper proposes a new localization method for indoor mobile robots. Using two cameras and one laser range finder on board a TRC mobile robot, the initial position and pose of the robot can be obtained by multisensor fusion and scene matching based on geometric hashing. No correspondence calculation and special pattern recognition are needed during the scene matching. This localization method can be implemented in five stages: 1) Model the indoor environment. Some selected indoor environment features are firstly modeled off-line into hashing tables. 2) Perform system calibration and information fusion from two cameras and the range finder. 3) Extract the vertical edge points corresponding to the horizontal scanning plane of the 2D laser range finder from the scene images and transform them into geometric invariants. 4) Perform scene matching and matching verification by geometric hashing and model back- projection method respectively. 5) Perform position and pose estimation by a least square fit method. Experimental results show that the accuracy and reliability of this localization method ar quite high.

The need for automating cargo management in a manufacture environment is urgent. Recent advance in big data and artificial intelligence areas has made it possible. As we known, most cargoes are transported using autonomously guided vehicles or manpower. Many autonomously guided vehicles follow paths that are predefined and usually marked by labels that can be easily found by vehicles. The use of predefined path for vehicle navigation greatly limits the deployment of autonomously guided vehicles. A few vehicles use high accuracy Light Laser Detection and Ranging for sensing and guidance. In this paper, a low-cost and precise method for mobile robots has been developed. Considering the high nonlinearity of UWB ranging data, a least square method is used to find optimal locations and a gradient decent approach is applied for fast convergence to the optimal results. The proposed method is also able to diagnose the ultra-wide band data and will throw away any data corrupted by noise. Therefore, the robustness is obtained.

This paper presents a global localization method for mobile robots based on the geometric information of structured indoor environments. With a global/local point cloud and projection map, lines are extracted from the projection maps using Hough transform. According to the directions of the obtained lines, the orientations of projection maps and point clouds are normalized. Next, the template matching algorithm is applied to the normalized global and local projection maps. Once coarse localization is completed, final accurate localization is achieved using the Iterative Closest Points (ICP) algorithm. Experimental results on several point clouds show that the proposed method can achieve high localization accuracy in real-time. The proposed method can be used for other global localization applications in structured environments.

The hybrid localization method of straight line matching, corner matching and odometry is proposed to resolve the difficult localization problems of mobile robot operating in family indoor environment. Localization of robot is a challenge problem which involves of determining robot positioning relative to objects in the environment. The system is implemented using National Instrutment Labview platform to gain satisfied real time performance with CompactRIO support. Robot detects environment using a 2D laser range finder with TCP/IP protocol. Line feature extraction process include area divided, iterative end point fit (IEPF) and a least square technique is introduced. Based on line feature, straight lines and corners as geometry features are obtained. The detected straight lines and corners are matched with the global straight lines and corners to obtain the robot position/orientation value. The odometry localization algorithm, straight line localization algorithm and corner localization algorithm are discussed. Different weight values are distributed to straight lines, corner and odometry according to the error model. As a result stable localization is achieved with position and orientation resolution as 50mm, 5 degree. A good performance for the method is also achieved with cycle time as 120ms. Experiment shows the effectiveness of the hybrid localization method.

For the issue of increased computational complexity to improve the positioning accuracy of robots leaving in the mobile robot localization method, this paper propose a new Monte Carlo localization algorithm. This method combinated the traditional particle filter algorithm with unscented Kalman filter, markovian Monte Carlo and reduced complexities level through dynamically updating the number of particles in particle collection and ensuring the accuracy of mobile robot localization. Simulation results show that the algorithm can not only inhibit the particle degradation and improve the positioning accuracy of the robot, but also in terms of computational complexity has increased significantly.

Simultaneous Localizatoin and Mapping (SLAM) is an active area of robot research and the location technology of mobile robots is a very critical issue in the filed of SLAM. Low cost and high performance can not be balanced in commercial robots and we plan to achieve better performance with low-cost sensors. This paper focuses on the study of complex indoor environment and a new method of point cloud matching and low-cost mobile robot localization is presented. Firstly, we change the data within a range of distance from laser to image by building grid maps and we consider each map has no scaling relation to others. Second, fast Fourier transformation is used to get the rotation angle. In order to get translation parameters, one-dimensional Fourier transformation is used to horizontal projection and vertical projection of map. We can build maps based on positioning results. Finally, we perform comparative experiments with other common methods.

This paper proposes a 3D localization method for an outdoor mobile robot. This method assesses the 3D position including the altitude information, which is impossible in the existing 2D localization method. In this method, the 3D position of the robot is predicted using an encoder and an inclination sensor. The predicted position is fused with the position information obtained from the DGPS and the digital compass using extended kalman filter to evaluate the 3D position of the robot. The experimental results showed that the proposed method can effectively evaluate the 3D position of the robot in a sloping environment. Moreover, this method was found to be more effective than the conventional 2D localization method even in the evaluation of the plane position where altitude information is unnecessary.

In augmented reality system, registration is always a problem which is difficult to resolve effectively, especially for marker less registration. Due to the placed marks for marked visual registration method, it would cause visual pollution, and the registration often fails when the mark is partially occluded. Registered quality directly affects the fusion effects of the virtuality and reality. This paper proposed a marker less tracking registration method combined natural feature points with affine reprojection. Firstly it extracted natural feature points from real scene with SIFT algorithm, then exploited KLT algorithm to track and match the effective feature points in each image, carried out the homography matrix of each frame with affine reprojection technology, and computed the real scene 3D registration for each frame image, so as to implement the tracking registration effectively. This method is also applicable to those cases of plane feature points being occluded. Index Terms - Augmented Reality, Markerless, Tracking, Registration

Following an investigation undertaken by the publisher, we have determined that this paper was accepted on the basis of a compromised peer review process. We hereby retract the paper. The corresponding author has been notified of the retraction. The retraction statement can be found here: https://doi.org/10.1520/JTE20269995. Aiming at the problem that traditional methods cannot meet the problem of synchronous positioning of multi-mobile robots in different motion states, a synchronous positioning method of multi-mobile robots based on visual simultaneous localization and mapping (SLAM) algorithm is proposed. Analyze the general model of SLAM problem, realize visual SLAM feature detection through scale invariant feature transform algorithm, and match image features based on active vision. A multi-mobile robot motion state estimation model is constructed, an octree model is used to construct a multi-robot synchronous localization map, and the camera pose tracking is realized by using the oriented fast accelerated segment test and rotated binary robust independent elementary features points that are not in the target bounding box. The three-dimensional points in the environment recovered by visual SLAM separate the background target and the moving target, and match the corresponding target in the previous frame through the feature to realize the synchronous positioning of the multi-mobile robot. The experimental results show that the method can realize the synchronous positioning of multi-robots under the state of multi-robot linear motion, curvilinear motion and mixed motion, and the positioning accuracy is high.

Aiming at the complex and relatively closed underground environment of coal mine where mobile robot work, this paper proposes an autonomous localization method for coal mine underground mobile robot based on multi-sensor fusion. A simple robot localization model based on turning is established, which replaces the traditional method of estimating complex ground parameters in coal mine tunnel. In order to improve locating accuracy, a multi-sensor fusion localization algorithm based on EKF+AMCL (Extended Kalman Filter+ Adaptive Monte Carlo Localization) is proposed. The EKF algorithm is used to eliminate the white Gaussian noise in the path angle signal of the robot. By the AMCL algorithm, the location and path angle of the robot is fused with Lidar and Map data to obtain real time location and path angle after corrected errors. The experiment shows that in the established robot turning tunnel environment, the high-precision autonomous localization of the robot can be achieved, with an average location error of 0.027m and an average path angle error of 2.23°.