Vision-based Navigation System Research Articles

Precision navigation system design for Moon penetrator

Purpose Penetrator plays an important role in the exploration of Moon and Mars. The navigation method is a key technology during the development of penetrator. To meet the high accuracy requirements of Moon penetrator, this paper aims to propose two kinds of navigation systems. Design/methodology/approach The line of sight of vision sensor between the penetrator and Moon orbiter could be utilized as the measurement during the navigation system design. However, the analysis of observability shows that the navigation system cannot estimate the position and velocity of penetrator, when the line of sight measurement is the only resource of information. Therefore, the Doppler measurement due to the relative motion between penetrator and the orbiter is used as the supplement. The other option is the relative range measurement between penetrator and the orbiter. The sigma-point Kalman Filtering is implemented to fuse the information from the vision sensor and Doppler or rangefinder. The observability of two navigation system is analyzed. Findings The sigma-point Kalman filtering could be used based on vision sensor and Doppler radar or laser rangefinder to give an accurate estimation of Moon penetrator position and velocity without increasing the payload of Moon penetrator or decreasing the estimation accuracy. However, the simulation result shows that the last method is better. The observability analysis also proves this conclusion. Practical implications Two navigation systems are proposed, and the simulations show that both systems can provide accurate estimation of states of penetrator. Originality/value Two navigation methods are proposed, and the observability of these navigation systems is analyzed. The sigma-point Kalman filtering is first introduced to the vision-based navigation system for Moon penetrator to provide precision navigation during the descent phase of Moon penetrator.

Vision-based positioning system

Conventional outdoor navigation systems are usually based on orbital satellites, e.g., global positioning system (GPS) and global navigation satellite system (GLONASS). The latest advances from wearable, e.g., BaiduEye and Google Glass, have enabled new approaches to leverage information from the surrounding environment. For example, they enable the change from passively receiving information to actively requesting information. Thus, such changes might inspire brand new application scenarios that were not possible before. In this work, we propose a vision-based navigation system based on wearable like Baidu Eye. We discuss the associated challenges and propose potential solutions for each of them. The system utilizes crowd sensing to collect and build a traffic signpost database for positioning reference. Then it leverages context information, such as cell identification (Cell ID), signal strength, and altitude combined with traffic sign detection and recognition to enable real-time positioning. A hybrid cloud architecture is proposed to enhance the capability of sensing devices (SD) to realize the proposed vision.

Vision-Based Real-Time Positioning and Autonomous Navigation System Using Mobile Robot in Indoor Environments

Research toward unmanned mobile robot navigation has gained significant importance in the last decade due to its potential applications in the location-based services industry. The increase in construction of large space indoor buildings has made difficulty for humans to operate within such environments. In this study, a mobile robot's indoor navigation algorithm is developed with vision cameras. Using two monocular cameras (one looking forward and one looking downward), the developed algorithms make use of the salient features of the environments to estimate rotational and translational motions for real-time positioning of the mobile robot. At the same time, an algorithm based on artificial landmark recognition is developed. The artificial landmark is shaped arrow based signboards with different colors representing different paths. These algorithms are integrated into a designed framework for mobile robot real-time positioning and autonomous navigation. Experiments are performed to validate the designed system using the mobile robot PIONEER P3-AT. The developed algorithm was able to detect and extract artificial landmark information up to 3 m distance for the mobile robot guidance. Experiment results show an average error of 0.167 m deviation from the ideal path, signified the good ability and performance of the development autonomous navigation algorithm.

Vision-Based Autonomous Underwater Vehicle Navigation in Poor Visibility Conditions Using a Model-Free Robust Control

This paper presents a vision-based navigation system for an autonomous underwater vehicle in semistructured environments with poor visibility. In terrestrial and aerial applications, the use of visual systems mounted in robotic platforms as a control sensor feedback is commonplace. However, robotic vision-based tasks for underwater applications are still not widely considered as the images captured in this type of environments tend to be blurred and/or color depleted. To tackle this problem, we have adapted thelαβcolor space to identify features of interest in underwater images even in extreme visibility conditions. To guarantee the stability of the vehicle at all times, a model-free robust control is used. We have validated the performance of our visual navigation system in real environments showing the feasibility of our approach.

Relative attitude and position estimation for a tumbling spacecraft

This paper presents a novel relative attitude and position estimation approach for a tumbling spacecraft. It is assumed that the tumbling chief spacecraft is in failure or out of control and there is no a priori rotation rate information. The Euler's rotational dynamics is used to propagate the chief angular velocity, and the unknown inertia parameter circumstance is also considered. The integrated sensor suit comprises a rate-integrating gyro and a vision-based navigation system on the deputy spacecraft. Two relative quaternions that map the chief Local Vertical Local Horizontal (LVLH) frame to the deputy and chief body frames are involved to construct the line-of-sight observations. Therefore, the assumption that the chief body frame coincides with its LVLH frame in the traditional algorithm can be released. The general relative equations of motion for eccentric orbits are used to describe the positional dynamics. An extended Kalman filter is derived to estimate the relative quaternions, relative position and velocity, deputy gyro bias, as well as chief angular velocity and inertia ratios. Simulation results verify the validity and feasibility of the proposed algorithm.

Image-Based Method for Determining Better Walking Strategies for Hexapods

An intelligent walking strategy is vital for multi-legged robots possessing no a priori information of an environment when traversing across discontinuous terrain. Six-legged robots outperform other multi-legged robots in static and dynamic stability. However, hexapods require careful planning to traverse across discontinuous terrain. A hexapod walking strategy can be accomplished using a vision-based navigation system to identify the surrounding environment. This paper presents an image-based technique to achieve better walking strategies for a hexapod walking on a special terrain containing irregular, restricted regions. The properties of the restricted regions were acquired beforehand by using reliable surveillance means. Moreover, simplified forward gaits, better rotational gaits, and adaptive gait selection strategies for walking on discontinuous terrain were proposed. The hexapod can effectively switch the gait sequences and types according to the environment involved. The boundary of standing zones can be successfully labelled by applying the greyscale erosion comprising a structuring element similar in shape and size to the foot tip of the hexapod. The experimental results demonstrated that the proposed image-based technique significantly improved the walking strategies of hexapods traversing on discontinuous terrain.

Unscented Kalman Filtering for Relative Spacecraft Attitude and Position Estimation

A novel relative spacecraft attitude and position estimation approach based on an Unscented Kalman Filter (UKF) is derived. The integrated sensor suite comprises the gyro sensors on each spacecraft and a vision-based navigation system on the slave spacecraft. In the traditional algorithm, an assumption that the master's body frame coincides with its Local Vertical Local Horizontal (LVLH) frame is made to construct the line-of-sight observations for convenience. To solve this problem, two relative quaternions that map the master's LVLH frame to the slave and master body frames are involved. The general relative equations of motion for eccentric orbits are used to describe the positional dynamics. The implementation equations for the UKF are derived. A modified version of the UKF is presented based on the averaging-quaternion algorithm. Simulation results indicate that the proposed filters provide more accurate estimates of relative attitude and position than the Extended Kalman Filter (EKF).

Hybrid vision-based navigation for mobile robots in mixed indoor/outdoor environments

In this paper we present a vision-based navigation system for mobile robots equipped with a single, off-the-shelf camera in mixed indoor/outdoor environments. A hybrid approach is proposed, based on the teach-and-replay technique, which combines a path-following and a feature-based navigation algorithm. We describe the navigation algorithms and show that both of them correct the robot’s lateral displacement from the intended path. After that, we claim that even though neither of the methods explicitly estimates the robot position, the heading corrections themselves keep the robot position error bound. We show that combination of the methods outperforms the pure feature-based approach in terms of localization precision and that this combination reduces map size and simplifies the learning phase. Experiments in mixed indoor/outdoor environments were carried out with a wheeled and a tracked mobile robots in order to demonstrate the validity and the benefits of the hybrid approach.

Cubature Kalman filtering for relative spacecraft attitude and position estimation

A novel relative spacecraft attitude and position estimation approach based on cubature Kalman filter is derived. The integrated sensor suit comprises the gyro sensors on each spacecraft and a vision-based navigation system on the deputy spacecraft. In the traditional algorithm, an assumption that the chief׳s body frame coincides with its Local Vertical Local Horizontal (LVLH) frame is made to construct the line-of-sight observations for convenience. To solve this problem, two relative quaternions that map the chief׳s LVLH frame to the deputy and chief body frames are involved. The general relative equations of motion for eccentric orbits are used to describe the positional dynamics. The implementation equations for the cubature Kalman filter are derived. Simulation results indicate that the proposed filter provides more accurate estimates of relative attitude and position over than the extended Kalman filter.

Image matching techniques for vision-based indoor navigation systems: a 3D map-based approach1

In this paper, we give an overview of image matching techniques for various vision-based navigation systems: stereo vision, structure from motion and map-based approach. Focused on map-based approach, which generally uses feature-based matching for mapping and localisation, and based on our early developed system, a performance analysis has been carried out and three major problems have been identified: inadequate geo-referencing and imaging geometry for mapping, vulnerability to drastic viewpoint changes and big percentage of mismatches at positioning. We introduce multi-image matching for mapping. By using affine-scale-invariant feature transform for viewpoint changes, the major improvement takes place on the epoch with large viewpoint changes. In order to deal with mismatches that were unable to be removed by Random Sample Consensus (RANSAC), we propose new method that use cross-correlation information to evaluate the quality of homography model and select the proper one. The conducted experiments have proved that such an approach can reduce the chances of mismatches being included by RANSAC and final positioning accuracy can be improved.

Structural similarity-based ranking of stereo algorithms for dynamic adaptation in real-time robot navigation

Real-time navigation of mobile robot in an unconstrained environment is a challenging area of research and it has several applications such as autonomous systems, military applications, security systems, etc. State-of-the-art stereo vision-based robot navigation system uses a specific stereo matching algorithm during complete navigation, which fails to handle radiometric variations such as change in illumination and exposure. The available illumination and exposure invariant stereo algorithms are computationally expensive and may not be suitable for real-time applications. This research ranked the existing correlation-based stereo algorithms for different real-time conditions by analysing its quality of the disparity using range structural similarity index measure. The proposed evaluation ranking table can be used to generate better disparity at each instance during the robot navigation.

Optical Navigation System for Pin-Point Lunar Landing

Abstract Major space agencies have an increasing interest in highly accurate (200 m) autonomous landing on the Moon. Inertial-only navigation is not compatible with this challenging requirement. The techniques currently investigated rely on vision-based navigation. A first approach consists in tracking features between sequences of images in order to measure the angular rate as well as the direction of the velocity vector of the spacecraft. A second approach aims at identifying image features using a geo-referenced on-board database to determine the attitude and the position of the spacecraft. However, existing algorithms are computationally prohibitive and have a limited robustness to varying illumination conditions and surface characteristics. This paper presents the development of an innovative autonomous vision-based navigation system addressing these problems. Numerical simulations have shown that this system is capable of estimating the position and velocity of the vehicle with an accuracy better than 100 m and 0.1 m/s respectively. This work is the result of a successful collaboration between the Universite de Sherbrooke, NGC Aerospace Ltd., Thales Alenia Space and the European Space Agency. The proposed system has been selected as the main navigation algorithm in three major research and development projects sponsored by European Space Agency and the Canadian Space Agency.

Pose estimation of an uncooperative spacecraft from actual space imagery

This paper addresses the preliminary design of a spaceborne monocular vision-based navigation system for on-orbit-servicing and formation-flying applications. The aim is to estimate the pose of a passive space resident object using its known three-dimensional model and single low-resolution two-dimensional images collected on-board the active spacecraft. In contrast to previous work, no supportive means are available on the target satellite (e.g., light emitting diodes) and no a-priori knowledge of the relative position and attitude is available (i.e., lost-in-space scenario). Three fundamental mechanisms – perceptual organisation, true perspective projection, and random sample consensus – are exploited to overcome the limitations of monocular passive optical navigation in space. The preliminary design is conducted and validated making use of actual images collected in the frame of the PRISMA mission at about 700 km altitude and 10 m inter-spacecraft separation.

Geometrical distortion integrated performance index for vision-based navigation system

This paper proposes weighted dilution of precision (WDOP) as an indicator of the accuracy of position and attitude in vision-based navigation. WDOP accurately represents the tendencies of navigational errors. It is obtained by weighted least squares. The weight is determined by the deployment of feature points and the geometrical distortion of the vision sensor. The performance of WDOP was verified by simulation. The values of the dilution of precision (DOP) and WDOP were computed and analyzed by comparison with the navigational errors. Additionally, a correlation test was used to determine how well they reflect the trends of the navigational errors. Simulation results showed that WDOP was strongly correlated with navigational errors, which makes it a parameter that can be used to determine the quality of a vision-based navigation system. The proposed WDOP can be used as a practical indicator of navigation performance.

LOW-COST NAVIGATION AND GUIDANCE SYSTEMS FOR UNMANNED AERIAL VEHICLES — PART 2: ATTITUDE DETERMINATION AND CONTROL

ABSTRACT This paper presents the second part of the research activity performed by Cranfield University to assess the potential of low-cost navigation sensors for Unmanned Aerial Vehicles (UAVs). This part focuses on carrier-phase Global Navigation Satellite Systems (GNSS) for attitude determination and control of small to medium size UAVs. Recursive optimal estimation algorithms were developed for combining multiple attitude measurements obtained from different observation points (i.e., antenna locations), and their efficiencies were tested in various dynamic conditions. The proposed algorithms converged rapidly and produced the required output even during high dynamics manoeuvres. Results of theoretical performance analysis and simulation activities are presented in this paper, with emphasis on the advantages of the GNSS interferometric approach in UAV applications (i.e., low cost, high data-rate, low volume/weight, low signal processing requirements, etc.). The simulation activities focussed on the AEROSONDE UAV platform and considered the possible augmentation provided by interferometric GNSS techniques to a low-cost and low-weight/volume integrated navigation system (presented in the first part of this series) which employed a Vision-Based Navigation (VBN) system, a Micro-Electro-Mechanical Sensor (MEMS) based Inertial Measurement Unit (IMU) and code-range GNSS (i.e., GPS and GALILEO) for position and velocity computations. The integrated VBN-IMU-GNSS (VIG) system was augmented using the inteferometric GNSS Attitude Determination (GAD) sensor data and a comparison of the performance achieved with the VIG and VIG/GAD integrated Navigation and Guidance Systems (NGS) is presented in this paper. Finally, the data provided by these NGS are used to optimise the design of a hybrid controller employing Fuzzy Logic and Proportional-Integral-Derivative (PID) techniques for the AEROSONDE UAV.

Development of Microscopic Hardness and Stiffness Investigation System with Microrobot 2nd Report, Vision Based Precise Navigation

A microsurface measurement system that is composed of the microrobot with the indenter and a vision based navigation system is proposed for investigating hardness and stiffness of such microparts. Here the tiny robot with the electromagnetic legs and the piezo elements incorporates with an electromagnetic driven microforce generator. This force generator can provide small forces up to 17 mN with 50 µN resolutions and push down the microindenter to the surface. The displacement of the indenter head can be also measured by the Linear Valuable Differential Transformer (LVDT) on machine. Thus, this mechanism can generate the small force and monitor the depth behaviour of the indenter during whole dwell time. Since the overall size of this mechanism is small enough to implement on the piezo-driven microrobot, the tiny robot with the microindenter is capable tomove precisely step by step with 1 µm per step so that the microindenter could be penetrated anywhere on the sample surface. With the help of an image processing technique, the vision based coordination system with the local close-up view and the overall global view has been developed to identify the locations of small robot and the indenter precisely within ±3 µm accuracy over the working range. In the experimental results, several results that the indentation load-depth characteristics of the unhealthy human tooth are measured automatically at the specified point are discussed.

Vision-aided Navigation for Autonomous Aircraft Based on Unscented Kalman Filter

A vision-aided navigation system for autonomous aircraft is described in this paper. The vision navigation of the aircraft to the known scence is performed with a camera fixed on the aircraft. The location and pose of the aircraft are estimated with the corresponding control points which can be detected in the images captured. The control points are selected according their saliency and are tracked in sequential images based on Fourier-Melline transform. The simulation model of the aircraft dynamics and vision-aided navigation system based on Matlab/Simulink is built.The unscented Kalman filter is used to fuse the aircraft state information provided by the vision system and the inertial navigation system. Simulation results show that the vision-based navigation system provides satisfactory results of both accuracy and reliability. DOI: http://dx.doi.org/10.11591/telkomnika.v11i2.2250 Full Text: PDF

Functioning methods and algorithms of color vision-based correlation-extremal aircraft navigation system

Existing high-accuracy aircraft navigation systems are based on inertial navigation systems integrated with aiding systems. An aiding system can be represented by an RGB color vision-based correlationextremal navigation system (CENS). The article describes the algorithm for consideration of air condition when making the color photos for reference images and the algorithm for synthesizing the reference images of one season using the color photos of the other season to be applied in color vision-based CENS.

Carrier-phase GNSS Attitude Determination and Control for Small Unmanned Aerial Vehicle Applications

As part of our recent research to assess the potential of low-cost navigation sensors for Unmanned Aerial Vehicle (UAV) applications, we investigated the potential of carrier-phase Global Navigation Satellite System (GNSS) for attitude determination and control of small size UAVs. Recursive optimal estimation algorithms were developed for combining multiple attitude measurements obtained from different observation points (i.e., antenna locations), and their efficiencies were tested in various dynamic conditions. The proposed algorithms converged rapidly and produced the required output even during high dynamics manoeuvres. Results of theoretical performance analysis and simulation activities are presented in this paper, with emphasis on the advantages of the GNSS interferometric approach in UAV applications (i.e., low cost, high data-rate, low volume/weight, low signal processing requirements, etc.). The simulation activities focussed on the AEROSONDE UAV platform and considered the possible augmentation provided by interferometric GNSS techniques to a low-cost and low-weight/volume integrated navigation system (presented in the first part of this series) which employed a Vision-Based Navigation (VBN) system, a Micro- Electro-Mechanical Sensor (MEMS) based Inertial Measurement Unit (IMU) and code-range GNSS (i.e., GPS and GALILEO) for position and velocity computations. The integrated VBN-IMU-GNSS (VIG) system was augmented using the inteferometric GNSS Attitude Determination (GAD)sensor data and a comparison of the performance achieved with the VIG and VIG/GAD integrated Navigation and Guidance Systems (NGS) is presented in this paper. Finally, the data provided by these NGS are used to optimise the design of a hybrid controller employing Fuzzy Logic and Proportional-Integral-Derivative (PID) techniques for the AEROSONDE UAV.

A Low-cost Vision Based Navigation System for Small Size Unmanned Aerial Vehicle Applications

A low cost navigation system based on Vision Based Navigation (VBN) and other avionics sensors is presented, which is designed for small size Unmanned Aerial Vehicle (UAV) applications. The main objective of our research is to design a compact, lightweight and relatively inexpensive system capable of providing the required navigation performance in all phases of flight of a small UAV, with a special focus on precision approach and landing, where Vision Based Navigation (VBN) techniques can be fully exploited in a multisensory integrated architecture. Various existing techniques for VBN are compared and the Appearance-based Navigation (ABN) approach is selected for implementation. Feature extraction and optical flow techniques are employed to estimate flight parameters such as roll angle, pitch angle, deviation from the runway and body rates. Additionally, we address the possible synergies between VBN, Global Navigation Satellite System (GNSS) and MEMS-IMU (Micro-Electromechanical System Inertial Measurement Unit) sensors and also the use of Aircraft Dynamics Models (ADMs) to provide additional information suitable to compensate for the shortcomings of VBN and MEMS-IMU sensors in high-dynamics attitude determination tasks. An Extended Kalman Filter (EKF) is developed to fuse the information provided by the different sensors and to provide estimates of position, velocity and attitude of the UAV platform in real-time. Two different integrated navigation system architectures are implemented. The first uses VBN at 20 Hz and GPS at 1 Hz to augment the MEMS-IMU running at 100 Hz. The second mode also includes the ADM (computations performed at 100 Hz) to provide augmentation of the attitude channel. Simulation of these two modes is performed in a significant portion of the AEROSONDE UAV operational flight envelope and performing a variety of representative manoeuvres (i.e., straight climb, level turning, turning descent and climb, straight descent, etc.). Simulation of the first integrated navigation system architecture (VBN/IMU/GPS) shows that the integrated system can reach position, velocity and attitude accuracies compatible with CAT-II precision approach requirements. Simulation of the second system architecture (VBN/IMU/GPS/ADM) also shows promising results since the achieved attitude accuracy is higher using the ADM/VBS/IMU than using VBS/IMU only. However, due to rapid divergence of the ADM virtual sensor, there is a need for frequent re-initialisation of the ADM data module, which is strongly dependent on the UAV flight dynamics and the specific manoeuvring transitions performed.