GPS-denied Environments Research Articles

A Path-Planning Method for Wall Surface Inspection Robot Based on Improved Genetic Algorithm

A wall surface inspection robot mainly relies on the inertial measurement unit and global positioning system (GPS) signal during intelligent wall surface inspection. The robot may encounter incorrect positioning under a GPS-denied environment, easily triggering safety accidents. In order to obtain a path suitable for the safe work of the robot wall surface inspection robot in a GPS-denied environment, a global path-planning method for wall surface inspection robots was proposed based on the improved generic algorithm (GA). The influencing factor for GPS signal strength was introduced into the heuristic function in path planning for GA to address the aforementioned problem. Using the PSO algorithm, GA was initialized and the influencing term of GPS signal was introduced into the fitness degree function so as to achieve point-to-point path planning of vertical wall surface inspection robot. Path angle and probability of intersection and variation was taken into account for better path planning capability. Finally, the simulation experiments were performed. The generated path using the improved GA was found to avoid the blind area of the GPS signal. The algorithm proposed has a good performance with average convergence times of 35.9 times and an angle of 55.88° in simple environment. Contrary to the traditional GA and PSO algorithm, the method showed advantages in terms of the convergence rate, path quality, path angle change, and algorithm stability. The research presented in this article is meaningful and relatively sufficient. The simulation test is also quite convincing. The proposed method was proved to be effective in global path planning for a wall surface inspection robot.

Towards Improved Inertial Navigation by Reducing Errors Using Deep Learning Methodology

Autonomous vehicles make use of an Inertial Navigation System (INS) as part of vehicular sensor fusion in many situations including GPS-denied environments such as dense urban places, multi-level parking structures, and areas with thick tree-coverage. The INS unit incorporates an Inertial Measurement Unit (IMU) to process the linear acceleration and angular velocity data to obtain orientation, position, and velocity information using mechanization equations. In this work, we describe a novel deep-learning-based methodology, using Convolutional Neural Networks (CNN), to reduce errors from MEMS IMU sensors. We develop a CNN-based approach that can learn from the responses of a particular inertial sensor while subject to inherent noise errors and provide near real-time error correction. We implement a time-division method to divide the IMU output data into small step sizes to make the IMU outputs fit the input format of the CNN. We optimize the CNN approach for higher performance and lower complexity that would allow its implementation on ultra-low power hardware such as microcontrollers. Our results show that we achieved up to 32.5% error improvement in straight-path motion and up to 38.69% error improvement in oval motion compared with the ground truth. We examined the performance of our CNN approach under various situations with IMUs of various performance grades, IMUs of the same type but different manufactured batch, and controlled, fixed, and uncontrolled vehicle motion paths.

Autonomous Source Term Estimation in Unknown Environments: From a Dual Control Concept to UAV Deployment

In the gas source search and localisation problem, the use of autonomous robots is of increasing interest due to their deployment speed and lack of human interaction with hazardous materials. This letter presents an aerial robotic platform for performing source term estimation of an unknown chemical release in a challenging a-priori unknown and GPS-denied environment. The proposed system forms the search strategy using the state-of-the-art control concept, dual control for exploitation and exploration, and realises such a function in the aforementioned challenging scenario using an RRT* based path planner. A novel downsampling process on the RRT* is also proposed that addresses the computational infeasibility of calculating the utility of a large number of sample states, whilst still maintaining sample state diversity. The proposed algorithm is tested in a high fidelity simulation environment under a number of configurations, and compared against competing algorithms. The system architecture is also brought forward into a bespoke UAV platform and experimentally tested in real-world conditions. The proposed system is shown to be capable of performing source term estimation robustly and efficiently, which provides a step forward in showing the real world application of previously academic functions.

Large-Scale Autonomous Flight With Real-Time Semantic SLAM Under Dense Forest Canopy

Semantic maps represent the environment using a set of semantically meaningful objects. This representation is storage-efficient, less ambiguous, and more informative, thus facilitating large-scale autonomy and the acquisition of actionable information in highly unstructured, GPS-denied environments. In this letter, we propose an integrated system that can perform large-scale autonomous flights and real-time semantic mapping in challenging under-canopy environments. We detect and model tree trunks and ground planes from LiDAR data, which are associated across scans and used to constrain robot poses as well as tree trunk models. The autonomous navigation module utilizes a multi-level planning and mapping framework and computes dynamically feasible trajectories that lead the UAV to build a semantic map of the user-defined region of interest in a computationally and storage efficient manner. A drift-compensation mechanism is designed to minimize the odometry drift using semantic SLAM outputs in real time, while maintaining planner optimality and controller stability. This leads the UAV to execute its mission accurately and safely at scale. Code is released at: https://github.com/KumarRobotics/kr_autonomous_flight and https://github.com/KumarRobotics/sloam.

Flight in GPS-denied environment: Autonomous navigation system for micro-aerial vehicle

Autonomous navigation in GPS-denied environments is a challenging task for MAV (Micro-Aerial Vehicle). That needs the MAV to do localization quickly and robustly while avoiding obstacles without the priori knowledge of environment. The limited onboard resources also restrict perceptive capability and computation. In this paper, a completed solution about platform design and software architecture is proposed. The perception system is proposed for localization and perceiving the environment. The motion planning and control system are presented for collision avoidance. Experimental testing reveals that the proposed solution can achieve fast and robust MAV autonomous navigation in GPS-denied environments.

Autonomous dynamic docking of UAV based on UWB-vision in GPS-denied environment

This paper studies the autonomous docking between an Unmanned Aerial Vehicle (UAV) and a Mobile Platform (MP) based on UWB and vision sensors. To solve this problem, an integrated estimation and control scheme is proposed, which is divided into three phases: hovering, approaching and landing. In the hovering phase, the velocity of the MP and relative position between the MP and UAV are estimated by using geometric tools and Cayley-Menger determinant based on ultra-wideband distance measurements; in the approaching phase, a recursive least squares optimization algorithm with a forgetting factor is proposed, which uses distance, displacement and MP’s velocity to estimate the relative position between the UAV and MP. With the estimated relative position, UAV can approach MP until reaching a distance such that MP is within the field of view of UAV; in the landing phase, the UWB measurement value and visual perception attitude are integrated with the UAV on-board navigation sensor of the UAV to perform the precision landing. Simulation and experiment results verify the effectiveness and feasibility of the proposed integrated navigation scheme.

System for multi-robotic exploration of underground environments CTU-CRAS-NORLAB in the DARPA Subterranean Challenge

We present a field report of the CTU-CRAS-NORLAB team from the Subterranean Challenge (SubT) organized by the Defense Advanced Research Projects Agency (DARPA). The contest seeks to advance technologies that would improve the safety and efficiency of search-andrescue operations in GPS-denied environments. During the contest rounds, teams of mobile robots have to find specific objects while operating in environments with limited radio communication, e.g., mining tunnels, underground stations or natural caverns. We present a heterogeneous exploration robotic system of the CTU-CRAS-NORLAB team, which achieved the third rank at the SubT Tunnel and Urban Circuit rounds and surpassed the performance of all other non-DARPA-funded teams. The field report describes the team’s hardware, sensors, algorithms and strategies, and discusses the lessons learned by participating at the DARPA SubT contest.

Robotic Precision Harvesting: Mapping, Localization, Planning and Control for a Legged Tree Harvester

This paper presents a system for autonomously conducting precision harvesting missions using a legged harvester. Precision tree harvesting removes some trees selectively, while leaving neighboring trees intact. Our robot performs the challenging task of navigation and tree grabbing in a confined, GPS-denied forest environment. We propose strategies for mapping, localization, planning, and control and integrate them into a fully autonomous system. The mission starts with a human mapping the area of interest using a detachable, custom sensor module. Subsequently, a human expert selects the specific trees for harvesting. The sensor module is then mounted on the machine and used for localization within the created map. A planning algorithm searches for both an approach-pose and a path in a single path planning problem. We design a path-following controller exploiting the legged harvester’s capabilities for negotiating rough terrain. Upon reaching the approach-pose, the machine grabs a tree with a general-purpose gripper. Our system has been tested in both emulated and natural forest settings. To the best of our knowledge, ours is the first robot to demonstrate such a level of autonomy on a full-size, hydraulic machine operating in a realistic environment.

Vision-Based 2D Navigation of Unmanned Aerial Vehicles in Riverine Environments with Imitation Learning

There have been many researchers studying how to enable unmanned aerial vehicles (UAVs) to navigate in complex and natural environments autonomously. In this paper, we develop an imitation learning framework and use it to train navigation policies for the UAV flying inside complex and GPS-denied riverine environments. The UAV relies on a forward-pointing camera to perform reactive maneuvers and navigate itself in 2D space by adapting the heading. We compare the performance of a linear regression-based controller, an end-to-end neural network controller and a variational autoencoder (VAE)-based controller trained with data aggregation method in the simulation environments. The results show that the VAE-based controller outperforms the other two controllers in both training and testing environments and is able to navigate the UAV with a longer traveling distance and a lower intervention rate from the pilots.

A Novel Polarized Skylight Navigation Model for Bionic Navigation With Marginalized Unscented Kalman Filter

Bionic navigation is an essential technology in GPS-denied environment for vehicle navigation. This paper combined strapdown inertial navigation system, polarized skylight sensors and odometer to design a new navigation model. In particular, a novel measurement model is developed based on polarized skylight. This model primarily utilizes the cross product to compute the error between the measured polarization vector and the theoretical polarization vector. It differs from the error computed by strapdown inertial navigation system in design vector measurement model. It deals with the problem of the directional ambiguity of polarization vector difference directly. Considering the measured polarization vector error due to computation and multiple scattering from atmospheric molecules, the states are augmented with the measured vector error to form a partially nonlinear, bionic integrated navigation model. To reduce the computation burden, marginalised unscented Kalman filter is presented to estimate the unknown states. The observability analysis method of piece-wise constant systems is applied to compute the observability matrix and singular value. Simulation results show that linear maneuver and angular maneuver can improve the degree of the observability of attitude and measured polarized vector error. Finally, experiments further verify the effectiveness of proposed models and method.

Autonomous aerial flight path inspection using advanced manufacturing techniques

SUMMARYRobotic systems have shown capabilities to perform inspection tasks in dangerous and difficult-to-access environments, such as those found in different components of power plants. However, most of the current robotic inspection technology is designed for specific components. Aerial robots, commonly termed as Drones, have raised an option to inspect a wider range of structural components. Nevertheless, current aerial inspecting technology still relies on a human pilot with limited line of sight, field of view and a reduced perception as the drone flies away, which prevents performing close-quarter inspection in intricate, structurally complex and GPS-denied environments. This work introduces offline inspection path generation methods based on robotics-integrated to manufacturing techniques. One method uses computer-aided manufacturing (CAM) techniques and the other an additive manufacturing (AM) approach to generate the flight path. That is to say, the drone would fly along the path described by a 3D (Three Dimensional) printer’s extruder or a CNC (Computer Numerical Control) machining tool, enabling to fly very close the structure even in physical structures with a complex geometry. Once the trajectories are generated, they are introduced for its validation in the Gazebo robotics simulator. Simulation results demonstrate the proper performance of the method and confirm that this approach can be used for close inspection of structuralcomponents.

Mobile Robot Circumnavigating an Unknown Target Using Only Range Rate Measurement

This brief proposes a bounded continuous control law to steer a mobile robot to approach and maintain a predefined distance from an unknown stationary target in a GPS-denied environment where the positions of the robot and the target are not available. The control algorithm utilizes only the derivative of the distance between the mobile robot and the target, that is, range rate measurement. We demonstrate that the controller can achieve global convergence regardless of the initial states of the mobile robot. Compared with the existing works, the proposed control algorithm is the simplest form (derivative-type controller) and easy to implement. Numerical simulations and real-time experiments are both performed to confirm the effectiveness of the proposed control scheme.

Enhanced Attitude and Altitude Estimation for Indoor Autonomous UAVs

In recent years the use of Unmanned Aerial Vehicles (UAVs) has considerably grown in the civil sectors, due to their high flexibility of use. Currently, two important key points are making them more and more successful in the civil field, namely the decrease of production costs and the increase in navigation accuracy. In this paper, we propose a Kalman filtering-based sensor fusion algorithm, using a low cost navigation platform that contains an inertial measurement unit (IMU), five ultrasonic ranging sensors and an optical flow camera. The aim is to improve navigation in indoor or GPS-denied environments. A multi-rate version of the Extended Kalman Filter is considered to deal with the use of heterogeneous sensors with different sampling rates, and the presence of non-linearities in the model. The effectiveness of the proposed sensor platform is evaluated by means of numerical tests on the dynamic flight simulator of a quadrotor. Results show high precision and robustness of the attitude estimation algorithm, with a reduced computational cost, being ready to be implemented on low-cost platforms.

GPS-Denied Navigation Aided by Synthetic Aperture Radar Using the Range-Doppler Algorithm

<h3>Abstract</h3> The need to successfully navigate in the absence of GNSS has grown in recent years. In particular, light aircraft such as UAVs are growing in popularity for a variety of applications vulnerable to GPS denial. The research presented here develops a GPS-denied navigation scheme for light aircraft employing images formed from a synthetic aperture radar system. Past research has explored the utility of radar telemetry in GPS-denied systems. This research advances previous work by exploiting radar images to obtain range and cross-range position measurements. Images are formed using the Range-Doppler Algorithm, an efficient image formation algorithm ideal for the sometimes limited processing packages available to light aircraft. An inertial navigation and radar processing system is implemented using both real and simulated radar images to aid in estimating an aircraft’s state in a GPS-denied environment. The results show that navigation in the absence of GPS using synthetic aperture radar is feasible with converging and bounded estimation errors.

Design, analysis and prototyping of a spherical drone for underground mines

This paper aims to provide performance optimisation of a multirotor drone that is designed for underground spaces. Currently, drones are being used efficiently for various applications in surface fieldwork projects. But using drones for underground spaces is challenging because of confined space, GPS-denied environment and various air properties in underground spaces. In this paper, first, the designed and manufactured drone is discussed. Then, to improve the drone performance for defined missions, a sensitivity analysis is done over the effective design parameters of the drone components and air density as an environmentally effective parameter. The sensitivity analysis over the flight simulations shows that drone performance increases by increasing air density, while the optimised number of propellers blades is presented. The results show that by increasing the battery capacity, the flight time increases but the overall performance of the drone decrease. Also, the drone would be able to carry about 450 grams payload and payloads more than 450 grams can significantly decrease flight time. Finally, the results of sensitivity analysis lead to optimise the drone performance efficiently for the designed mission in underground mines.

Problems solved during AEROBOT-2021 UAV Challenge

This is a report on Institute of Control Sciences (ICS) team participation in AEROBOT-2021 UAV challenge. It is a contest of autonomous navigation by multirotor unmanned aerial vehicles in an unfamiliar GPS-denied environment. AEROBOT was held for the 3rd time, and ICS team successfully finished the course. We describe the approaches we used to solve competition tasks. In particular, the visual navigation that works in the actual arena is selected. Our flight strategies work reliably even with visual odometry errors. An original black lane segmentation algorithm using exponential context histogram and priority direction estimation is presented. The on-site adaptation of neural network models for detecting objects has been tested in practice.

A Fast Indoor Positioning Using a Knowledge-Distilled Convolutional Neural Network (KD-CNN)

Fingerprint-based indoor positioning systems (F-IPS) may provide inexpensive solutions to GPS-denied environments. Most F-IPSs adopt traditional machine learning for position prediction, resulting in low accuracy. Deep neural networks (DNN) were recently employed for F-IPSs to minimize positioning errors. Nevertheless, a DNN-IPS fails to guarantee high accuracy in dynamic environments as it is sensitive to changes in the input data. A convolutional neural network (CNN) is recommended to replace DNN due to its capability to learn the overall topology of fingerprinting images and capture highly abstract features. Due to the convolution process and image representation, CNN-IPS incurs prohibitive storage and computational requirement for implementation on resource-limited devices. This paper incorporates knowledge distillation (KD) into CNN-IPS to distil knowledge from large deep CNNs into small CNNs. The pretrained teacher network uses the soft probability output where the score vector from the trained network is converted into a probability distribution, which is softened by the temperature hyperparameter, leading to a more simplified model. Based on the numerical results, KD-CNN-IPS manifests better localization performance where 79.84% of the positioning errors are within 2 meters while its testing time is only 79.68% of that of the teacher model. Compared to the CNN-IPS, KD-CNN-IPS with precisely the same architecture and size could achieve a performance improvement of 13.65% in terms of the average positioning error.

Autonomous Path Travel Control of Mobile Robot Using Internal and External Camera Images in GPS-Denied Environments

In this study, we developed a system for calculating the relative position and angle between a mobile robot and a marker using information such as the size of the marker of the internal camera of the mobile robot. Using this information, the mobile robot runs autonomously along the path given by the placement of the marker. In addition, we provide a control system that can follow a trajectory using information obtained by recognizing the mobile robot when reflected in an external camera using deep learning. The proposed method can easily achieve autonomous path travel control for mobile robots in environments where GPS cannot be received. The effectiveness of the proposed system is demonstrated under several actual experiments.

Structure from motion with planar homography estimation: a real-time low-bandwidth, high-resolution variant for aerial reconnaissance

We propose a new algorithm variant for Structure from Motion (SfM) to enable real-time image processing of scenes imaged by aerial drones. Our new SfM variant runs in real-time at 4 Hz equating to an 80× computation time speed-up compared to traditional SfM and is capable of a 90% size reduction of original video imagery, with an added benefit of presenting the original two-dimensional (2D) video data as a three-dimensional (3D) virtual model. This opens many potential applications for a real-time image processing that could make autonomous vision–based navigation possible by completely replacing the need for a traditional live video feed. The 3D reconstruction that is generated comes with the added benefit of being able to generate a spatially accurate representation of a live environment that is precise enough to generate global positioning system (GPS) coordinates from any given point on an imaged structure, even in a GPS-denied environment.

2-D UAV navigation solution with LIDAR sensor under GPS-denied environment

Unmanned aerial vehicle (UAV) is widely used by many industries these days such as militaries, agriculture, and surveillance. However, one of the main challenges of UAV is navigating through an environment where global positioning system (GPS) is being denied. The main purpose of this paper is to find a solution for UAV to be able to navigate in a GPS denied surrounding without affecting the drone flight performance. There are two ways to overcome these challenges such as using visual odometry (VO) or by using simultaneous localization and mapping (SLAM). However, VO has a drawback because camera sensors require good lighting which will affect the performance of the UAV when it is navigating through a low light intensity environment. Hence, in this paper 2-D SLAM will be use as a solution to help UAV to navigate under a GPS-denied environment with the help of a light detection and ranging (LIDAR) sensor which known as a LIDAR-based SLAM. This is because SLAM can help UAVs to localize itself and map the surrounding of the environment. The concept and idea of this paper will be fully simulated using MATLAB, where the drone navigation will be simulated in MATLAB to extract LIDAR data and to use the LIDAR data to carry out SLAM via pose graph optimization. Besides, the contribution to this research work has also identified that in pose graph optimization, the loop closure threshold and loop closure radius play an important role. The loop closure threshold can affect the accuracy of the trajectory of the drone and the accuracy of mapping the environment as compared to ground truth. On the other hand, the loop closure search radius can increase the processing speed of obtaining the data via pose graph optimization. The main contribution to this research work is shown that the processing speed can increase up to 45 % and the accuracy of the trajectory of the drone and the mapped surrounding is quite accurate as compared to ground truth.