Academic literature on the topic 'Obstacle Aided Navigation'

Natural environments are often filled with obstacles and disturbances. Traditional navigation and planning approaches normally depend on finding a traversable “free space” for robots to avoid unexpected contact or collision. We hypothesize that with a better understanding of the robot–obstacle interactions, these collisions and disturbances can be exploited as opportunities to improve robot locomotion in complex environments. In this article, we propose a novel obstacle disturbance selection (ODS) framework with the aim of allowing robots to actively select disturbances to achieve environment-aided locomotion. Using an empirically characterized relationship between leg–obstacle contact position and robot trajectory deviation, we simplify the representation of the obstacle-filled physical environment to a horizontal-plane disturbance force field. We then treat each robot leg as a “disturbance force selector” for prediction of obstacle-modulated robot dynamics. Combining the two representations provides analytical insights into the effects of gaits on legged traversal in cluttered environments. We illustrate the predictive power of the ODS framework by studying the horizontal-plane dynamics of a quadrupedal robot traversing an array of evenly-spaced cylindrical obstacles with both bounding and trotting gaits. Experiments corroborate numerical simulations that reveal the emergence of a stable equilibrium orientation in the face of repeated obstacle disturbances. The ODS reduction yields closed-form analytical predictions of the equilibrium position for different robot body aspect ratios, gait patterns, and obstacle spacings. We conclude with speculative remarks bearing on the prospects for novel ODS-based gait control schemes for shaping robot navigation in perturbation-rich environments.

As difficult vision-based tasks like object detection and monocular depth estimation are making their way in real-time applications and as more light weighted solutions for autonomous vehicles navigation systems are emerging, obstacle detection and collision prediction are two very challenging tasks for small embedded devices like drones. We propose a novel light weighted and time-efficient vision-based solution to predict Time-to-Collision from a monocular video camera embedded in a smartglasses device as a module of a navigation system for visually impaired pedestrians. It consists of two modules: a static data extractor made of a convolutional neural network to predict the obstacle position and distance and a dynamic data extractor that stacks the obstacle data from multiple frames and predicts the Time-to-Collision with a simple fully connected neural network. This paper focuses on the Time-to-Collision network’s ability to adapt to new sceneries with different types of obstacles with supervised learning.

The task of organizing an informative route as an optimal spline trajectory of a moving object with an assessment of the informativeness of the map-aided navigation standard is completed. From the perspective of the planning approach, an optimal geometric path, passing through pre-determined iconic intermediate points, taking into account the avoidance of navigational hazards as “obstacle spots” has been formed. Within the framework of the strategy of informative planning of the spline path, the actuality of solving the problem of synthesizing the optimal trajectory in two variants is noted: by the methods of B-splines and classical polynomial interpolations as the implementation of the tactics of a mobile object movement in a conflict environment. A comparative characteristic of two alternative algorithms for solving the problem, specifying the advantages and disadvantages of each option, is given. As a demonstration of the practical applicability of the interpolation approach, the spline trajectory of an illustrative example in route of map-aided navigation is designed against the background of a contour map of isolines. Emphasis is placed on the possibility of forming the shape of the navigation isosurface due to the effective use of the curvature of the spline trajectory as a reproductive template for constructing an axonometric projection. A forecast about the trends of the possible use of a separate optimal trajectory of the object movement directly for the construction of the informative field profile of any degree of complexity is made. A hypothesis about the feasibility of practical use of chaotic architecture of spline gradients for effective planning of the optimal trajectory is put forward. The fan of spline gradient vectors with a personal orientation in the direction of the maximum change in the navigation function on each segment of the piecewise polyline of the path in the vessel routing procedure is considered. The issue of ensuring the possibility of making a coordinated decision on the vessel management by personnel due to the automated formation of spline trajectories in real time with synchronous representation of geometric computer support to the watch assistant, which allows us to offer integration of the tasks under consideration into the cloud-based intelligent technology of “augmented reality” is formulated.

Powered wheelchairs have enhanced the mobility and quality of life of people with special needs. The next step in the development of powered wheelchairs is to incorporate sensors and electronic systems for new control applications and capabilities to improve their usability and the safety of their operation, such as obstacle avoidance or autonomous driving. However, autonomous powered wheelchairs require safe navigation in different environments and scenarios, making their development complex. In our research, we propose, instead, to develop contactless control for powered wheelchairs where the position of the caregiver is used as a control reference. Hence, we used a depth camera to recognize the caregiver and measure at the same time their relative distance from the powered wheelchair. In this paper, we compared two different approaches for real-time object recognition using a 3DHOG hand-crafted object descriptor based on a 3D extension of the histogram of oriented gradients (HOG) and a convolutional neural network based on YOLOv4-Tiny. To evaluate both approaches, we constructed Miun-Feet—a custom dataset of images of labeled caregiver’s feet in different scenarios, with backgrounds, objects, and lighting conditions. The experimental results showed that the YOLOv4-Tiny approach outperformed 3DHOG in all the analyzed cases. In addition, the results showed that the recognition accuracy was not improved using the depth channel, enabling the use of a monocular RGB camera only instead of a depth camera and reducing the computational cost and heat dissipation limitations. Hence, the paper proposes an additional method to compute the caregiver’s distance and angle from the Powered Wheelchair (PW) using only the RGB data. This work shows that it is feasible to use the location of the caregiver’s feet as a control signal for the control of a powered wheelchair and that it is possible to use a monocular RGB camera to compute their relative positions.

People living with a degenerative retinal disease such as retinitis pigmentosa are oftentimes faced with difficulties navigating in crowded places and avoiding obstacles due to their severely limited field of view. The study aimed to assess the potential of different patterns of eye movement (scanning patterns) to (i) increase the effective area of perception of participants with simulated retinitis pigmentosa scotoma and (ii) maintain or improve performance in visual tasks. Using a virtual reality headset with eye tracking, we simulated tunnel vision of 20° in diameter in visually healthy participants (n = 9). Employing this setup, we investigated how different scanning patterns influence the dynamic field of view—the average area over time covered by the field of view—of the participants in an obstacle avoidance task and in a search task. One of the two tested scanning patterns showed a significant improvement in both dynamic field of view (navigation 11%, search 7%) and collision avoidance (33%) when compared to trials without the suggested scanning pattern. However, participants took significantly longer (31%) to finish the navigation task when applying this scanning pattern. No significant improvements in search task performance were found when applying scanning patterns.

This paper designed a multi-information fusion algorithm after analysis information from vision sensors and radar sensors. This algorithm used D-S evidence theory to fuse the information of vision sensors and radar sensors to judge the front obstacles, and a final decision is made by the distance information provided by radar to decide whether give the driver corresponding warning. It also designed a critical vehicle distance, which can change according to relative distance and relative velocity. The test results show that this algorithm can give warning information correctly and greatly decrease the uncertainty, thus satisfying the requirement of car aided navigation system. At a resolution of 320×480, the identifying speed of this algorithm can reach 62.5ms/F which satisfied the requirement of real-time of car navigation.

GPS signal loss is a major issue when the navigation system of rovers is based solely on GPS for outdoor navigation rendering the rover stuck in the mid of the road in case of signal loss. In this study, a low-cost IMU aided GPS-based navigation system for Ackermann Steered mobile robots is presented and tested to cater to the issue of GPS signal loss along. GPS path is selected and fed using the android application which provides real-time location tracking of the rover on the map embedded into the application. System utilizes Arduino along with the node MCU, compass, IMU, Rotary encoders, and an Ackermann steered rover. Controller processes the path file, compares its current position with the path coordinates and navigates using inertial sensor aided navigation algorithm, avoiding obstacles to reach its destination. IMU measures the distance traveled from each path point, and in case of signal loss, it makes the rover move for the remaining distance in the direction of destination point. Rover faced a sinusoidal motion due to the steering, so PID was implemented. The system was successfully tested on the IST premises and finds its application in the delivery trolley, institutional delivery carts, and related applications.

An H∞ controller combined with an Artificial Potential Field Method (APFM) was applied to seabed navigation for Autonomous Underwater Vehicles (AUVs), aimed particularly at obstacle avoidance and bottom-following operations in the vertical plane. Depth control and altitude control prevented the AUV from colliding with the sea bottom or with obstacles and prevented the AUV from diving beyond its maximum depth limit when bottom following. Simulation and laboratory trials with various seabed contours indicated that with the H∞ controller, the AUV was able to safely reach appointed destinations without collisions. Tests also showed that the H∞ controller was robust and suppressed interference, hence ensuring the precision of its navigation control. The proposed H∞ controller combined with the APFM has thus been proved to be both feasible and effective.

The use of blind-aided navigating systems has become very essential in the 4th industrial revolution. The essence of this is to improve navigating autonomy for the blind. Although, navigation challenges are not of serious concern to people with eye defects. However, this is a major concern for blind people due to the time-consuming navigation process. These limitations necessitate the use of walking sticks, dogs, or people to navigate their path. This project provides a blind navigation system based on Light Detection and Ranging Based Navigation System for Blind People (LiDAR). The components of this fabricated device involve a charging system, a LiDAR sensor, a microcontroller, a calling stick, two buzzers, a switch vibration motor with an RF transmitter, and a receiver to give object detection and real-time assistance to the blind. The mode of its operation is by detecting obstacles along the route of a blind person and giving a notification via the buzzer and the vibration motor. Also, during emergencies, the buzzer rings and informs the visually impaired person using the device. On the other hand, the microcontroller and other modules in the device have a continuous ongoing relationship which is beneficial to the user. Another appreciable benefit of this device is the ability to recharge the battery component, the device is of low-cost, quick, simple-to-use, and novel solution for the blind.

Bathymetry is a subset of hydrography, aimed at measuring the depth of waterbodies and waterways. Measurements are taken inter alia to detect natural obstacles or other navigational obstacles that endanger the safety of navigation, to examine the navigability conditions, anchorages, waterways and other commercial waterbodies, and to determine the parameters of the safe depth of waterbodies in the vicinity of ports, etc. Therefore, it is necessary to produce precise and reliable seabed maps, so that any hazards that may occur, particularly in shallow waterbodies, can be prevented, including the high dynamics of hydromorphological changes. This publication is aimed at developing a concept of an innovative autonomous unmanned system for bathymetric monitoring of shallow waterbodies. A bathymetric and topographic system will use autonomous unmanned aerial and surface vehicles to study the seabed relief in the littoral zone (even at depths of less than 1 m), in line with the requirements set out for the most stringent International Hydrographic Organization (IHO) order—exclusive. Unlike other existing solutions, the INNOBAT system will enable the coverage of the entire surveyed area with measurements, which will allow a comprehensive assessment of the hydrographic and navigation situation in the waterbody to be conducted.

Autonomous underwater vehicles are rapidly becoming more common and developed for both military as well as civilian applications such as; mine countermeasures or in the field of geoscience. One of the main issues these vehicles are facing today is the navigation capability. Since they operate in an underwater environment, modern navigation strategies such as GPS and radar are not available due to the quick attenuation of the high frequency signals. The most commonly used navigation strategy is the dead-reckoning navigation which estimates the location of the vehicle using sensor data and numerical integration. The disadvantage with this technique is the accumulative errors during longer operation times. Terrain-aided navigation has been extensively researched as of late and deemed as a viable approach to compensate for the errors when using dead-reckoning navigation. This type of navigation uses a predefined map of the bathymetry and estimates the position of the vehicle on the map using a particle filter and bathymetry measurements from the sonars attached to the vehicle. This project is focused on finding the most suitable auxiliary sonar for the underwater vehicle which enables accurate terrainaided navigation with regards to physical limitations such as the weight, resolution, power consumption and accuracy of the sonar. The sonars to be tested are a single-beam echo sounder, multiple singe-beam echo sounders, multi-beam echo sounder and a 2.5D sonar. The four different sonars are tested with the terrain-aided navigation algorithm where a vehicle simulator developed by Saab is used for the simulations. The results show that the multiple single-beam echo sounder configuration is the most suitable solution due to its high accuracy during the navigation and its low impact on the system dynamics and power consumption. The chosen sonar configuration is later integrated together with a behavior-fusion algorithm to achieve reliable obstacle avoidance capabilities for the vehicle. The algorithm uses a weighted mean-value of the sonar beams to identify approaching obstacles as well as a bathymetry derivative estimation to make the system more adaptive to its surroundings. The vehicle was able to detect incoming obstacles ahead and to the sides of the vehicle up to 17 seconds earlier compared to when only using a doppler velocity log.
Autonoma undervattensfarkoster håller just nu på att slå igenom på bred front inom både militära och civila tillämpningar såsom; minjakt eller inom geovetenskap. Ett av de största problemen denna typ av farkoster har idag är förmågan att säkert navigera under uppdragen. Då de används i en undervattensmiljö så kan inte moderna navigationsmetoder såsom GPS eller radar användas då högfrekventa signaler dämpas snabbt. Den vanligaste typen av navigation är död räkning, vilket estimerar farkostens position genom användning av sensor data och numerisk integration. Nackdelen med denna typ av teknik är de ackumulerade felen som uppstår under längre körningar. På senare tid har terrängnavigering varit i fokus inom forskningen och anses vara en bra lösning för att kompensera för positionsfelen som uppstår vid navigation med död räkning. Denna typ av navigation använder sig av en fördefinierad karta över batymetrin och estimerar farkostens position relativt kartan med hjälp av ett partikel filter och batymetrimätningar från sonarer moneterade på farkosten. Detta projekt fokuserar på att hitta den mest passade sonaren för farkosten som gör det möjligt att använda terrängnavigering med hög precision då fysiska aspekter såsom, vikt, upplösning, strömförbrukning och noggranheten av sonaren tas till hänsyn. Sonarerna som skall testas består av; ett enstråligt ekolod, flera enstråliga ekolod, ett flerstråligt ekolod och en 2.5D sonar. De fyra sonarerna testas med terrängnavigeringsalgoritmen, där en simulator av farkosten utvecklad av Saab används som grund för simuleringarna. Resultaten visar att användningen av flera enstråliga ekolod är den mest passade lösningen då den resulterar i väldigt god navigationsförmåga med låg inverkan på farkostens strömförbrukning och dynamik jämfört med de andra alternativen. Den valda sonaren används sedan tillsammans med en behaviour-fusion baserad algoritm för att farkosten skall uppnå pålitlig hinderundvikning. Algoritmen använder sig av ett viktat medelvärde av sonarstrålarna för att identifiera närmade hinder tillsammans med en uppskattning av batymetrins derivata för att få en mer adaptiv hinderundvikning. Farkosten kunde med hjälp av behaviour-fusion algoritmen och de flerstråliga ekoloden identifiera närmande objekt upp till 17 sekunder tidigare än om bara en doppler velocity log användes.

Most micro air vehicles rely heavily on reliable GPS measurements for proper estimation and control, and therefore struggle in GPS-degraded environments. When GPS is not available, the global position and heading of the vehicle is unobservable. This dissertation establishes the theoretical and practical advantages of a relative navigation framework for MAV navigation in GPS-degraded environments. This dissertation explores how the consistency, accuracy, and stability of current navigation approaches degrade during prolonged GPS dropout and in the presence of heading uncertainty. Relative navigation (RN) is presented as an alternative approach that maintains observability by working with respect to a local coordinate frame. RN is compared with several current estimation approaches in a simulation environment and in hardware experiments. While still subject to global drift, RN is shown to produce consistent state estimates and stable control. Estimating relative states requires unique modifications to current estimation approaches. This dissertation further provides a tutorial exposition of the relative multiplicative extended Kalman filter, presenting how to properly ensure observable state estimation while maintaining consistency. The filter is derived using both inertial and body-fixed state definitions and dynamics. Finally, this dissertation presents a series of prolonged flight tests, demonstrating the effectiveness of the relative navigation approach for autonomous GPS-degraded MAV navigation in varied, unknown environments. The system is shown to utilize a variety of vision sensors, work indoors and outdoors, run in real-time with onboard processing, and not require special tuning for particular sensors or environments. Despite leveraging off-the-shelf sensors and algorithms, the flight tests demonstrate stable front-end performance with low drift. The flight tests also demonstrate the onboard generation of a globally consistent, metric, and localized map by identifying and incorporating loop-closure constraints and intermittent GPS measurements. With this map, mission objectives are shown to be autonomously completed.

Micro aerial vehicles and other autonomous systems have the potential to truly transform life as we know it, however much of the potential of autonomous systems remains unrealized because reliable navigation is still an unsolved problem with significant challenges. This dissertation presents solutions to many aspects of autonomous navigation. First, it presents ROSflight, a software and hardware architure that allows for rapid prototyping and experimentation of autonomy algorithms on MAVs with lightweight, efficient flight control. Next, this dissertation presents improvments to the state-of-the-art in optimal control of quadrotors by utilizing the error-state formulation frequently utilized in state estimation. It is shown that performing optimal control directly over the error-state results in a vastly more computationally efficient system than competing methods while also dealing with the non-vector rotation components of the state in a principled way. In addition, real-time robust flight planning is considered with a method to navigate cluttered, potentially unknown scenarios with real-time obstacle avoidance. Robust state estimation is a critical component to reliable operation, and this dissertation focuses on improving the robustness of visual-inertial state estimation in a filtering framework by extending the state-of-the-art to include better modeling and sensor fusion. Further, this dissertation takes concepts from the visual-inertial estimation community and applies it to tightly-coupled GNSS, visual-inertial state estimation. This method is shown to demonstrate significantly more reliable state estimation than visual-inertial or GNSS-inertial state estimation alone in a hardware experiment through a GNSS-GNSS denied transition flying under a building and back out into open sky. Finally, this dissertation explores a novel method to combine measurements from multiple agents into a coherent map. Traditional approaches to this problem attempt to solve for the position of multiple agents at specific times in their trajectories. This dissertation instead attempts to solve this problem in a relative context, resulting in a much more robust approach that is able to handle much greater intial error than traditional approaches.

Computer-Aided techniques have been deployed more commonly in recent years to assist with surgical procedures, particularly in the case of minimally invasive surgeries. Arthroscopy, as one of the most prevailing minimally invasive surgical procedures, increases surgical complexity due to the loss of joint visibility, but has many advantages. More obstacles are encountered during hip arthroscopy, given the tight socket-joint hip anatomy. Therefore, computer-aided techniques could be used to ease such difficulties during hip arthroscopy.

Three-dimensional sensing is a vital field in mobile robotic applications. This work proposes an application of a Time-of-Flight 3D Photonic Mixer Device (PMD) camera for the navigation of an Omni-directional mobile robot. The 3D PMD camera enables real time distance detection as well as the capturing of grayscale images. In our framework, the application of the 3D PMD camera is aimed at solving the problem of environmental perception in mobile robotics. In this paper, we present the development of a MATLAB-based kit for the control of an Omni-directional mobile robot supported by a data acquisition board. The communication interface of the camera, used to close the system’s control loop, has been also developed. We further present results of different experiments including online obstacle detection and avoidance. In addition, an adaptive pose determination for the robot is proposed.