Vision-based Localization System Research Articles

MINI UAV-BASED LITTER DETECTION ON RIVER BANKS

Abstract. Most of the anthropic pollution arriving to seas and oceans is carried by rivers, leading to a dramatic impact on the aquatic flora and fauna. Hence, several strategies have already been considered to reduce the use of plastic (and non biodegradable) objects, fostering recycling, detect litter in the environment, and catch it. This work aims at implementing a litter detection approach usable also in urban areas, hence considering a mini-UAV (Unmanned Aerial Vehicle) in order to reduce the issues related to flight restrictions. The implemented strategy exploits a high resolution map of the area of interest, a properly trained deep learning litter object detector, and a vision based localization system, which allows to remarkably reduce the positioning error of the UAV navigation system, in order to provide estimates of the litter object positions with an accuracy at decimeter level for objects not too far from locations recognizable in the map.

Vision-Based Localization System Suited to Resident Underwater Vehicles

In recent years, we have seen significant interest in the use of permanently deployed resident robotic vehicles for commercial inspection, maintenance and repair (IMR) activities. This paper presents a concept and demonstration, through offshore trials, of a low-cost, low-maintenance, navigational marker that can eliminate drift in vehicle INS solution when the vehicle is close to the IMR target. The subsea localisation marker system is fixed on location on the resident field asset and is used in on-vehicle machine vision algorithms for pose estimation and facilitation of a high-resolution world coordinate frame registration with a high refresh rate. This paper presents evaluation of the system during trials in the North Atlantic Ocean during January 2019. System performances and propagation of position error is inspected and estimated, and the effect of intermittent visual based position update to Kalman filter and onboard INS solution is discussed. The paper presents experimental results of the commercial state-of-the-art inertial navigation system operating in the pure inertial mode for comparison.

Vision-Based Localization and Mapping System for Navigating Autonomous Agricultural Vehicle

Vision-Based Localization and Mapping System for Navigating Autonomous Agricultural Vehicle

Leveraging experience for robust, adaptive nonlinear MPC on computationally constrained systems with time-varying state uncertainty

This paper presents a robust-adaptive nonlinear model predictive control (MPC) technique that leverages past experiences to achieve tractability on computationally constrained systems. We propose a robust extension of the Experience-driven Predictive Control (EPC) algorithm via a Gaussian belief propagation strategy that computes an uncertainty set, bounding the evolution of the system state in the presence of time-varying state uncertainty. This uncertainty set is used to tighten the constraints in the predictive control formulation via a chance-constrained approach, thereby providing a probabilistic guarantee of constraint satisfaction. The parameterized form of the controllers produced by EPC coupled with online uncertainty estimates ensures that this robust constraint satisfaction property persists, even as the system switches controllers and experiences variations in the uncertainty model. We validate the online performance and robust constraint satisfaction of the proposed Robust EPC algorithm through a series of trials with a simulated ground robot and three experimental platforms: (1) a small quadrotor aerial robot executing aggressive maneuvers in wind with degraded state estimates, (2) a skid-steer ground robot equipped with a laser-based localization system, and (3) a hexarotor aerial robot equipped with a vision-based localization system.

Indoor Pedestrian Localization With a Smartphone: A Comparison of Inertial and Vision-Based Methods

Indoor pedestrian navigation systems are increasingly needed in various types of applications. However, such systems are still face many challenges. In addition to being accurate, a pedestrian positioning system must be mobile, cheap, and lightweight. Many approaches have been explored. In this paper, we take the advantage of sensors integrated in a smartphone and their capabilities to develop and compare two low-cost, hands-free, and handheld indoor navigation systems. The first one relies on embedded vision (smartphone camera), while the second option is based on low-cost smartphone inertial sensors (magnetometer, accelerometer, and gyroscope) to provide a relative position of the pedestrian. The two associated algorithms are computationally lightweight, since their implementations take into account the restricted resources of the smartphone. In the experiment conducted, we evaluate and compare the accuracy and repeatability of the two positioning methods for different indoor paths. The results obtained demonstrate that the vision-based localization system outperforms the inertial sensor-based positioning system.

Leader-Follower Formation for Nonholonomic Mobile Robots: Discrete-Time Approach

This paper presents a novel solution for the classical leader-follower formation problem considering the case of nonholonomic mobile robots. A formation control strategy is proposed in a discrete-time context by considering the exact discrete-time discretization of the non-linear continuous-time kinematic model of the vehicle. The geometric formation of the robots allows us to derive an alternative model that describes the time evolution of the relative distance and angle between the robots. These variables are obtained in real-time by a vision-based localization system on board, in which the follower robot is equipped with a Kinect device, together with a recognition board mounted on the leader robot. The boundedness of the relative position error is formally proven by considering a feedback law that is delayed by one sampling period of time. Numerical simulations and real-time experiments are presented to verify the performance of the control strategy.

A multi-vision sensor-based fast localization system with image matching for challenging outdoor environments

A sensor-based vision localization system is one of the most essential technologies in computer vision applications like an autonomous navigation, surveillance, and many others. Conventionally, sensor-based vision localization systems have three inherent limitations, These include, sensitivity to illumination variations, viewpoint variations, and high computational complexity. To overcome these problems, we propose a robust image matching method to provide invariance to the illumination and viewpoint variations by focusing on how to solve these limitations and incorporate this scheme into the vision-based localization system. Based on the proposed image matching method, we design a robust localization system that provides satisfactory localization performance with low computational complexity. Specifically, in order to solve the problem of illumination and viewpoint, we extract a key point using a virtual view from a query image and the descriptor based on the local average patch difference, similar to HC-LBP. Moreover, we propose a key frame selection method and a simple tree scheme for fast image search. Experimental results show that the proposed localization system is four times faster than existing systems, and exhibits better matching performance compared to existing algorithms in challenging environments with difficult illumination and viewpoint conditions.

Trajectory-based visual localization in underwater surveying missions.

We present a new vision-based localization system applied to an autonomous underwater vehicle (AUV) with limited sensing and computation capabilities. The traditional EKF-SLAM approaches are usually expensive in terms of execution time; the approach presented in this paper strengthens this method by adopting a trajectory-based schema that reduces the computational requirements. The pose of the vehicle is estimated using an extended Kalman filter (EKF), which predicts the vehicle motion by means of a visual odometer and corrects these predictions using the data associations (loop closures) between the current frame and the previous ones. One of the most important steps in this procedure is the image registration method, as it reinforces the data association and, thus, makes it possible to close loops reliably. Since the use of standard EKFs entail linearization errors that can distort the vehicle pose estimations, the approach has also been tested using an iterated Kalman filter (IEKF). Experiments have been conducted using a real underwater vehicle in controlled scenarios and in shallow sea waters, showing an excellent performance with very small errors, both in the vehicle pose and in the overall trajectory estimates.

A VISION-BASED LOCATION POSITIONING SYSTEM VIA AUGMENTED REALITY: AN APPLICATION IN HUMANOID ROBOT NAVIGATION

In this paper, we present a vision-based localization system using mobile augmented reality (MAR) and mobile audio augmented reality (MAAR) techniques, applicable to both humans and humanoid robots navigation in indoor environments. In the first stage, we propose a system that recognizes the location of a user from the image sequence of an indoor environment using its onboard camera. The location information is added to the user's view in the form of 3D objects and audio sounds with location information and navigation instruction content via augmented reality (AR). The location is recognized by using the prior knowledge about the layout of the environment and the location of the AR markers. The image sequence can be obtained using a smart phone's camera and the marker detection, 3D object placement and audio augmentation will be performed by the phone's operating processor and graphical/audio modules. Using this system will majorly reduce the hardware complexity of such navigation systems, as it replaces a system consisting of a mobile PC, wireless camera, head-mounted displays (HMD) and a remote PC with a smart phone with camera. In the second stage, the same algorithm is employed as a novel vision-based autonomous humanoid robot localization and navigation approach. The proposed technique is implemented on a humanoid robot NAO and improves the robot's navigation and localization performance previously done using an extended Kalman filter (EKF) by presenting location-based information to the robot through different AR markers placed in the robot environment.

A low cost vision based localization system using fiducial markers

This paper investigates a mobile robot self-localization system based on fiducial markers which are placed on the ceiling. Recently there have been many articles related to path planning and coordination of mobile agents within an unknown environment. These algorithms demand dedicated (and often expensive) hardware that is capable of handling complex tasks in real time. In this paper, we have explored the possibility to realize an inexpensive and simple navigation system, based on passive fiducial markers, which are able to guide an autonomous mobile robot along a predefined path. Fiducial markers provide not only improved performance in runtime, but also much better identification and localization. Marker design based on circle shape is presented. Using a low-cost webcam and appropriate marker detection algorithm three features are determined: robot position, new movement direction and marker ID. Experimental results are presented at the end of the paper.