The fusion of visual and inertial cues has become popular in robotics due to the complementary nature of the two sensing modalities. While most fusion strategies to date rely on filtering schemes, the visual robotics community has recently turned to non-linear optimization approaches for tasks such as visual Simultaneous Localization And Mapping (SLAM), following the discovery that this comes with significant advantages in quality of performance and computational complexity. Following this trend, we present a novel approach to tightly integrate visual measurements with readings from an Inertial Measurement Unit (IMU) in SLAM. An IMU error term is integrated with the landmark reprojection error in a fully probabilistic manner, resulting to a joint non-linear cost function to be optimized. Employing the powerful concept of keyframes we partially marginalize old states to maintain a bounded-sized optimization window, ensuring real-time operation. Comparing against both vision-only and loosely-coupled visual-inertial algorithms, our experiments confirm the benefits of tight fusion in terms of accuracy and robustness

Mobile robot localization and mapping in unknown environments is a fundamental requirement for effective autonomous navigation. Three different approaches to localization and mapping are presented. Each is based on data collected from a robot using a dense range scanner to generate a planar representation of the surrounding environment. This externally sensed range data is then overlayed and correlated to estimate the robot's position and build a map. The three approaches differ in the choice of representation of the range data, but all achieve improvements over prior work using detailed sensor modeling and rigorous bookkeeping of the modeled uncertainty in the estimation processes. In the first approach, the raw range data points collected from two different positions are individually weighted and aligned to estimate the relative robot displacement. In the second approach, line segment features are extracted from the raw point data and are used as the basis for efficient and robust global map construction and localization. In the third approach, a new multi-scale data representation is introduced. New methods of localization and mapping are developed, taking advantage of this multi-scale representation to achieve significant improvements in computational complexity. A central focus of all three approaches is the determination of accurate and robust solutions to the data association problem, which is critical to the accuracy of any sensor-based localization and mapping method. Experiments using data collected from a Sick LMS-200 laser scanner illustrate the effectiveness of the algorithms and improvements over prior work. All methods are capable of being run in real time on a mobile robot, and can be used to support fully autonomous navigation applications.</p

Localization and Tracking are very interesting functionalities that can benefit a number of applications. Despite the large number of algorithms and technologies that have been proposed in this context, the literature still lacks a widely accepted solution, capable of cutting a tradeoff between service quality (i.e., localization accuracy) and device/architecture cost and complexity. In this paper, we tackle the problem from a different and rather new perspective: we investigate how the localization accuracy of nodes can be ameliorated by opportunistically exchanging localization information among nodes that occasionally happen to be in proximity. To this end, we define a simple though accuratemodel of the opportunistic interaction and we develop a technique based on Linear Matrix Inequalities and barycentric algorithm in order to localize a user node that is completely unaware of its position. We analyse the opportunistic localization performance for different settings of the design parameters as duty–cycle and error modeling taking into account accuracy, degradation in time, correlation among consecutive estimation

Many generic position-estimation algorithms are vulnerable to ambiguity introduced by nonunique landmarks. Also, the available high-dimensional image data is not fully used when these techniques are extended to vision-based localization. This paper presents the landmark matching, triangulation, reconstruction, and comparison (LTRC) global localization algorithm, which is reasonably immune to ambiguous landmark matches. It extracts natural landmarks for the (rough) matching stage before generating the list of possible position estimates through triangulation. Reconstruction and comparison then rank the possible estimates. The LTRC algorithm has been implemented using an interpreted language, onto a robot equipped with a panoramic vision system. Empirical data shows remarkable improvement in accuracy when compared with the established random sample consensus method. LTRC is also robust against inaccurate map data

Assistive robotic applications require systems capable of interaction in the human world, a workspace which is highly dynamic and not always predictable. Mobile assistive devices face the additional and complex problem of when and if intervention should occur; therefore before any trajectory assistance is given, the robotic device must know where it is in real-time, without unnecessary disruption or delay to the user requirements. In this paper, we demonstrate a novel robust method for determining room identification from floor features in a real-time computational frame for autonomous and assistive robotics in the human environment. We utilize two inexpensive sensors: an optical mouse sensor for straightforward and rapid, texture or pattern sampling, and a four color photodiode light sensor for fast color determination. We show how data relating floor texture and color obtained from typical dynamic human environments, using these two sensors, compares favorably with data obtained from a standard webcam. We show that suitable data can be extracted from these two sensors at a rate 16 times faster than a standard webcam, and that these data are in a form which can be rapidly processed using readily available classification techniques, suitable for real-time system application. We achieved a 95% correct classification accuracy identifying 133 rooms’ flooring from 35 classes, suitable for fast coarse global room localization application, boundary crossing detection, and additionally some degree of surface type identification

Efficient localisation is a highly desirable property for an autonomous navigation system. Weightless neural networks offer a real-time approach to robotics applications by reducing hardware and software requirements for pattern recognition techniques. Such networks offer the potential for objects, structures, routes and locations to be easily identified and maps constructed from fused limited sensor data as information becomes available. We show that in the absence of concise and complex information, localisation can be obtained using simple algorithms from data with inherent uncertainties using a combination of Genetic Algorithm techniques applied to a Weightless Neural Architecture

Multi-robot cooperative localization serves as an essential task for a team of mobile robots to work within an unknown environment. Based on the real-time laser scanning data interaction, a robust approach is proposed to obtain optimal multi-robot relative observations using the Metric-based Iterative Closest Point (MbICP) algorithm, which makes it possible to utilize the surrounding environment information directly instead of placing a localization-mark on the robots. To meet the demand of dealing with the inherent non-linearities existing in the multi-robot kinematic models and the relative observations, a robust extended H∞ filtering (REHF) approach is developed for the multi-robot cooperative localization system, which could handle non-Gaussian process and measurement noises with respect to robot navigation in unknown dynamic scenes. Compared with the conventional multi-robot localization system using extended Kalman filtering (EKF) approach, the proposed filtering algorithm is capable of providing superior performance in a dynamic indoor environment with outlier disturbances. Both numerical experiments and experiments conducted for the Pioneer3-DX robots show that the proposed localization scheme is effective in improving both the accuracy and reliability of the performance within a complex environment.This work was supported inpart by the National Natural Science Foundation of China under grants 61075094, 61035005 and 61134009

This paper presents a localization method for a mobile robot equipped with only low-cost ultrasonic sensors. Correlation-based Hough scan matching was used to obtain the robot&apos;s pose without any predefined geometric features. A local grid map and a sound pressure model of ultrasonic sensors were used to acquire reliable scan results from uncertain and noisy ultrasonic sensor data. The robot&apos;s pose was measured using correlation-based Hough scan matching, and the covariance was calculated. Localization was achieved by fusing the measurements from scan matching with the robot&apos;s motion model through the extended Kalman filter. Experimental results verified the performance of the proposed localization method in a real home environment. (c) 2012 Taylor & Francis and The Robotics Society of JapanX1189sciescopu

The goal of this thesis is to create a program, that will be receiving measurements from robot's sensors, provide sensor fusion and estimate position of robot based on this fusion. For solving I used knowledge about probabilistic robotics, robotic operating system, information fusion, filtering especially extended Kalman filter and robot localization. I created an application of extended Kalman filter as a result. Filter listen to messages from robot sensors, providing a sensor fusion and estimating position of the robot in environment. Filter can receive measurements from multiple sources. The estimated states have proven themselves reasonably accurate for successful robot localization in space

The guidance of a vehicle in an outdoor setting is typically implemented using a Real Time Kinematic Global Positioning System (RTK-GPS) potentially enhanced by auxiliary sensors such as electronic compasses, rotation encoders, gyroscopes, and vision systems. Since GPS does not function in an indoor setting where educational competitions are often held, an alternative guidance system was developed. This article describes a guidance method that contains a laser-based localization system, which uses a robot-borne single laser transmitter spinning in a horizontal plane at an angular velocity up to 81 radians per second. Sensor arrays positioned in the corners of a flat rectangular table with dimensions of 1.22 m × 1.83 m detected the laser beam passages. The relative time differences among the detections of the laser passages gave an indication of the angles of the sensors with respect to the laser beam transmitter on the robot. These angles were translated into Cartesian coordinates. The guidance of the robot was implemented using a uni-directional wireless serial connection and position feedback from the localization system. Three experiments were conducted to test the system: 1) the accuracy of the static localization system was determined while the robot stood still. In this test the average error among valid measurements was smaller than 0.3 %. However, a maximum of 3.7 % of the measurements were invalid due to several causes. 2) The accuracy of the guidance system was assessed while the robot followed a straight line. The average deviation from this straight line was 3.6 mm while the robot followed a path with a length of approximately 0.9 m. 3) The overall performance of the guidance system was studied while the robot followed a complex path consisting of 33 sub-paths. The conclusion was that the system worked reasonably accurate, unless the robot came in close proximity