Approach For Autonomous Navigation Research Articles

Digital twin‐based multi‐objective autonomous vehicle navigation approach as applied in infrastructure construction

AbstractThe widespread adoption of autonomous vehicles has generated considerable interest in their autonomous operation, with path planning emerging as a critical aspect. However, existing road infrastructure confronts challenges due to prolonged use and insufficient maintenance. Previous research on autonomous vehicle navigation has focused on determining the trajectory with the shortest distance, while neglecting road construction information, leading to potential time and energy inefficiencies in real‐world scenarios involving infrastructure development. To address this issue, a digital twin‐embedded multi‐objective autonomous vehicle navigation is proposed under the condition of infrastructure construction. The authors propose an image processing algorithm that leverages captured images of the road construction environment to enable road extraction and modelling of the autonomous vehicle workspace. Additionally, a wavelet neural network is developed to predict real‐time traffic flow, considering its inherent characteristics. Moreover, a multi‐objective brainstorm optimisation (BSO)‐based method for path planning is introduced, which optimises total time‐cost and energy consumption objective functions. To ensure optimal trajectory planning during infrastructure construction, the algorithm incorporates a real‐time updated digital twin throughout autonomous vehicle operations. The effectiveness and robustness of the proposed model are validated through simulation and comparative studies conducted in diverse scenarios involving road construction. The results highlight the improved performance and reliability of the autonomous vehicle system when equipped with the authors’ approach, demonstrating its potential for enhancing efficiency and minimising disruptions caused by road infrastructure development.

Autonomous Robotic Navigation Approach Using Deep Q-Network Late Fusion and People Detection-Based Collision Avoidance

In this work, we propose an approach for the autonomous navigation of mobile robots using fusion the of sensor data by a Double Deep Q-Network with collision avoidance by detecting moving people via computer vision techniques. We evaluate two data fusion methods for the proposed autonomous navigation approach: Interactive and Late Fusion strategy. Both are used to integrate mobile robot sensors through the following sensors: GPS, IMU, and an RGB-D camera. The proposed collision avoidance module is implemented along with the sensor fusion architecture in order to prevent the autonomous mobile robot from colliding with moving people. The simulation results indicate a significant impact on the success of completing the proposed mission by the mobile robot with the fusion of sensors, indicating a performance increase (success rate) of ≈27% in relation to navigation without sensor fusion. With the addition of moving people in the environment, deploying the people detection and collision avoidance security module has improved about the success rate by 14% when compared to that of the autonomous navigation approach without the security module.

A safe reinforcement learning approach for autonomous navigation of mobile robots in dynamic environments

AbstractWhen deploying mobile robots in real‐world scenarios, such as airports, train stations, hospitals, and schools, collisions with pedestrians are intolerable and catastrophic. Motion safety becomes one of the most fundamental requirements for mobile robots. However, until now, efficient and safe robot navigation in such dynamic environments is still an open problem. The critical reason is that the inconsistency between navigation efficiency and motion safety is greatly intensified by the high dynamics and uncertainties of pedestrians. To face the challenge, this paper proposes a safe deep reinforcement learning algorithm named Conflict‐Averse Safe Reinforcement Learning (CASRL) for autonomous robot navigation in dynamic environments. Specifically, it first separates the collision avoidance sub‐task from the overall navigation task and maintains a safety critic to evaluate the safety/risk of actions. Later, it constructs two task‐specific but model‐agnostic policy gradients for goal‐reaching and collision avoidance sub‐tasks to eliminate their mutual interference. Then, it further performs a conflict‐averse gradient manipulation to address the inconsistency between two sub‐tasks. Finally, extensive experiments are performed to evaluate the superiority of CASRL. Simulation results show an average 8.2% performance improvement over the vanilla baseline in eight groups of dynamic environments, which is further extended to 13.4% in the most challenging group. Besides, forty real‐world experiments fully illustrated that the CASRL could be successfully deployed on a real robot.

Quality-Aware Autonomous Navigation with Dynamic Path Cost for Vision-Based Mapping toward Drone Landing

This article presents a novel autonomous navigation approach that is capable of increasing map exploration and accuracy while minimizing the distance traveled for autonomous drone landings. For terrain mapping, a probabilistic sparse elevation map is proposed to represent measurement accuracy and enable the increasing of map quality by continuously applying new measurements with Bayes inference. For exploration, the Quality-Aware Best View (QABV) planner is proposed for autonomous navigation with a dual focus: map exploration and quality. Generated paths allow for visiting viewpoints that provide new measurements for exploring the proposed map and increasing its quality. To reduce the distance traveled, we handle the path-cost information in the framework of control theory to dynamically adjust the path cost of visiting a viewpoint. The proposed methods handle the QABV planner as a system to be controlled and regulate the information contribution of the generated paths. As a result, the path cost is increased to reduce the distance traveled or decreased to escape from a low-information area and avoid getting stuck. The usefulness of the proposed mapping and exploration approach is evaluated in detailed simulation studies including a real-world scenario for a packet delivery drone.

Cubature H∞ Filter based Simultaneous Localization and Mapping

For many real-real applications, autonomous navigation is necessary for many vehicles, including ground, underwater, and aerial vehicles. Simultaneous Localization and Mapping (SLAM) is an important approach for autonomous navigation in unknown environments and/or GPS-denied areas. In SLAM, by using the sensors embedded in the vehicle, the vehicle simultaneously builds a map and localise in that map. In this paper, a solution to the SLAM is proposed using cubature H∞filter (CH∞F). CH∞F is a derivative-free filter, which can deal with highly nonlinear and Gaussian or non-Gaussian systems. It has the desired features of both the cubature Kalman filter and H∞filter, deals with nonlinear and uncertain sensor dynamics. In this paper, CH∞F has been proposed for the SLAM application. Different simulation scenarios are considered and compared.

An autonomous navigation approach for unmanned vehicle in off-road environment with self-supervised traversal cost prediction

An autonomous navigation approach for unmanned vehicle in off-road environment with self-supervised traversal cost prediction

Multilayer Model-Aided INS Based on Real-Time Corrected Vehicle Dynamics Model

Vehicle dynamics model-aided inertial navigation system (INS) is an effective approach for autonomous navigation when the Global Navigation Satellite System fails, However, the accuracy of the vehicle dynamics model (VDM) will decrease, or even diverge, in low-speed and high-dynamic situations, which leads to the performance degradation of the whole system. The traditional VDM-aided INS uses the velocity information from VDM to estimate and correct INS errors, but most previous research ignored that the acceleration and angular velocity from INS are more accurate than VDM. To reach the full potential of INS and VDM, this paper proposed a three-layer structure for the VDM-aided INS. In the proposed method, the navigation information is divided into the sensor layer, system layer I, and system layer II. In the sensor layer, VDM is corrected by INS through the strong tracking cubature Kalman filter, and the chi-square test is adopted to avoid redundant computation. In the system layer I, INS is aided by the corrected velocity information from VDM. In the system layer II, the position is updated with the velocity and attitude that are corrected in the system layer I. As the result, VDM and INS can be corrected in different layers simultaneously, and the accuracy and robustness of the whole system can be improved. The simulation experiment proved that, with the sensor layer filter, the velocity errors of VDM can be reduced by 35% in low-speed and high-dynamic situations, and reduced by 20% in the normal situation. Then, a 31-minute field test shows that the maximum horizontal position error of the proposed method is 241.8 m, while those of two traditional VDM-aided INS methods are more than 500 m.

Serious game self-regulation using human-like agents to visualize students engagement base on crowd

Nowadays, the emergence of artificial intelligent (AI) technology for games has been advancely developed. A serious game is a technology employing AI to create a virtual environment in a serious gamification strategy. This research describes AI based virtual classrooms to adopt proper strategies and focusing on maintaining and increasing student engagement by encouraging self-regulation behavior at the learning process. The self-regulation behavior describes student's ability to direct their own learning to achieve learning targets on a path full of obstacles. By employing a human-like agent to visualize student engagement, this visualization aims to provide human-like experiences for users to comprehend student behavior. A reciprocal velocity obstacles (RVO)-based crowd behavior is employed to visualize student engagement. RVO is an autonomous navigation approach for directing the achievement of agents target. The human-like agents behave in various ways to reach the goal points depending on the performances and the obstacles before them. We employ our method in an investigation of students' learning activities in a pedagogically-centered learning environment at Universitas Islam Negeri (UIN) Walisongo, Semarang, Indonesia. The results demonstrate the best scenario changes along with the performances and obstacles faced to reach the goal points as well as the learning target.

Ground Testing of Digital Terrain Models to Prepare for OSIRIS-REx Autonomous Vision Navigation Using Natural Feature Tracking

The OSIRIS-REx (Origins, Spectral Interpretation, Resource Identification, and Security–Regolith Explorer) spacecraft collected a sample from the asteroid Bennu in 2020. This achievement leveraged an autonomous optical navigation approach called Natural Feature Tracking (NFT). NFT provided spacecraft state updates by correlating asteroid surface features rendered from previously acquired terrain data with images taken by the onboard navigation camera. The success of NFT was the culmination of years of preparation and collaboration to ensure that feature data would meet navigation requirements. This paper presents the findings from ground testing performed prior to the spacecraft's arrival at Bennu, in which synthetic data were used to develop and validate the technical approach for building NFT features. Correlation sensitivity testing using synthetic models of Bennu enabled the team to characterize the terrain properties that worked well for feature correlation, the challenges posed by smoother terrain, and the impact of imaging conditions on correlation performance. The team found that models constructed from image data by means of stereophotoclinometry (SPC) worked better than those constructed from laser altimetry data, except when test image pixel sizes were more than a factor of 2 smaller than those of the images used for SPC, and when topography was underrepresented and resulted in incorrect shadows in rendered features. Degradation of laser altimetry data related to noise and spatial sampling also led to poor correlation performance. Albedo variation was found to be a key contributor to correlation performance; topographic data alone were insufficient for NFT.

Passive Landmark Geometry Optimization and Evaluation for Reliable Autonomous Navigation in Mining Tunnels Using 2D Lidars.

Autonomous navigation in mining tunnels is challenging due to the lack of satellite positioning signals and visible natural landmarks that could be exploited by ranging systems. Solutions requiring stable power feeds for locating beacons and transmitters are not accepted because of accidental damage risks and safety requirements. Hence, this work presents an autonomous navigation approach based on artificial passive landmarks, whose geometry has been optimized in order to ensure drift-free localization of mobile units typically equipped with lidar scanners. The main contribution of the approach lies in the design and optimization of the landmarks that, combined with scan matching techniques, provide a reliable pose estimation in modern smoothly bored mining tunnels. A genetic algorithm is employed to optimize the landmarks’ geometry and positioning, thus preventing that the localization problem becomes ill-posed. The proposed approach is validated both in simulation and throughout a series of experiments with an industrial skid-steer CAT 262C robotic excavator, showing the feasibility of the approach with inexpensive passive and low-maintenance landmarks. The results show that the optimized triangular and symmetrical landmarks improve the positioning accuracy by 87.5% per 100 m traveled compared to the accuracy without landmarks. The role of optimized artificial landmarks in the context of modern smoothly bored mining tunnels should not be understated. The results confirm that without the optimized landmarks, the localization error accumulates due to odometry drift and that, contrary to the general intuition or belief, natural tunnel features alone are not sufficient for unambiguous localization. Therefore, the proposed approach ensures grid-based SLAM techniques can be implemented to successfully navigate in smoothly bored mining tunnels.

An autonomous navigation approach for unmanned vehicle in outdoor unstructured terrain with dynamic and negative obstacles

AbstractAt present, the study on autonomous unmanned ground vehicle navigation in an unstructured environment is still facing great challenges and is of great significance in scenarios where search and rescue robots, planetary exploration robots, and agricultural robots are needed. In this paper, we proposed an autonomous navigation method for unstructured environments based on terrain constraints. Efficient path search and trajectory optimization on octree map are proposed to generate trajectories, which can effectively avoid various obstacles in off-road environments, such as dynamic obstacles and negative obstacles, to reach the specified destination. We have conducted empirical experiments in both simulated and real environments, and the results show that our approach achieved superior performance in dynamic obstacle avoidance tasks and mapless navigation tasks compared to the traditional 2-dimensional or 2.5-dimensional navigation methods.

Realtime autonomous navigation in V-Rep based static and dynamic environment using EKF-SLAM

<p>Localization in an autonomous mobile robot allows it to operate autonomously in an unknown and unpredictable environment with the ability to determine its position and heading. Simultaneous localization and mapping (SLAM) are introduced to solve the problem where no prior information about the environment is available either static or dynamic to achieve standard map-based localization. The primary focus of this research is autonomous mobile robot navigation using the extended Kalman filter (EKF)-SLAM environment modeling technique which provides higher accuracy and reliability in mobile robot localization and mapping results. In this paper, EKF-SLAM performance is verified by simulations performed in a static and dynamic environment designed in V-REP i.e., 3D Robot simulation environment. In this work SLAM problem of two-wheeled differential drive robot i.e., Pioneer 3-DX in indoor static and dynamic environment integrated with Laser range finder i.e., Hokuyo URG-04LX- UG01, LIDAR, and Ultrasonic sensors is solved. EKF-SLAM scripts are developed using MATLAB that is linked to V-REP via remote API feature to evaluate EKF-SLAM performance. The reached results confirm the EKF- SLAM is a reliable approach for real-time autonomous navigation for mobile robots in comparison to other techniques.</p>

A Hybrid Approach for Autonomous Collision-Free UAV Navigation in 3D Partially Unknown Dynamic Environments

In the past decades, unmanned aerial vehicles (UAVs) have emerged in a wide range of applications. Owing to the advances in UAV technologies related to sensing, computing, power, etc., it has become possible to carry out missions autonomously. A key component to achieving this goal is the development of safe navigation methods, which is the main focus of this work. A hybrid navigation approach is proposed to allow safe autonomous operations in three-dimensional (3D) partially unknown and dynamic environments. This method combines a global path planning algorithm, namely RRT-Connect, with a reactive control law based on sliding mode control to provide quick reflex-like reactions to newly detected obstacles. The performance of the suggested approach is validated using simulations.

An Approach of Social Navigation Based on Proxemics for Crowded Environments of Humans and Robots.

Nowadays, mobile robots are playing an important role in different areas of science, industry, academia and even in everyday life. In this sense, their abilities and behaviours become increasingly complex. In particular, in indoor environments, such as hospitals, schools, banks and museums, where the robot coincides with people and other robots, its movement and navigation must be programmed and adapted to robot–robot and human–robot interactions. However, existing approaches are focused either on multi-robot navigation (robot–robot interaction) or social navigation with human presence (human–robot interaction), neglecting the integration of both approaches. Proxemic interaction is recently being used in this domain of research, to improve Human–Robot Interaction (HRI). In this context, we propose an autonomous navigation approach for mobile robots in indoor environments, based on the principles of proxemic theory, integrated with classical navigation algorithms, such as ORCA, Social Momentum, and A*. With this novel approach, the mobile robot adapts its behaviour, by analysing the proximity of people to each other, with respect to it, and with respect to other robots to decide and plan its respective navigation, while showing acceptable social behaviours in presence of humans. We describe our proposed approach and show how proxemics and the classical navigation algorithms are combined to provide an effective navigation, while respecting social human distances. To show the suitability of our approach, we simulate several situations of coexistence of robots and humans, demonstrating an effective social navigation.

Indoor Localization and Navigation based on Deep Learning using a Monocular Visual System

Now-a-days, computer systems are important for artificial vision systems to analyze the acquired data to realize crucial tasks, such as localization and navigation. For successful navigation, the robot must interpret the acquired data and determine its position to decide how to move through the environment. This paper proposes an indoor mobile robot visual-localization and navigation approach for autonomous navigation. A convolutional neural network and background modeling are used to locate the system in the environment. Object detection is based on copy-move detection, an image forensic technique, extracting features from the image to identify similar regions. An adaptive threshold is proposed due to the illumination changes. The detected object is classified to evade it using a control deep neural network. A U-Net model is implemented to track the path trajectory. The experiment results were obtained from real data, proving the efficiency of the proposed algorithm. The adaptive threshold solves illumination variation issues for object detection.

A Novel Implementation of a Color-Based Detection and Tracking Algorithm for an Autonomous Hexacopter

The increasing popularity of Unmanned Aerial Vehicles (UAVs), and the availability of associated supporting technologies such as sensors, development toolkits, and programming libraries have expanded the application areas of UAVs via customization of its functions and features. This technological trend has encouraged hobbyists to hack into off-the-shelf commercially available drones to serve different functions with minimal additional costs. In this paper, we present the practicality of using readily available off-the-shelf components in expanding a drone’s functionality by designing and implementing a real-time object-following autonomous hexacopter equipped with an on-board camera used as a primary sensor; this vision-based approach for autonomous navigation addresses the limitations of GPS-based navigation in an indoor environment. The system uses the CIE L*a*b* color space to perform color-based detection and tracking, and employs ultrasonic sensor information to avoid obstacles. In all the trials, the hexacopter was able to follow the targets 90.24% of the time, with guaranteed latency of at most two minutes.

A fast learning approach for autonomous navigation using a deep reinforcement learning method

Deep reinforcement learning-based methods employ an ample amount of computational power that affects the learning process. This paper proposes a novel approach to speed up the training process and improve the performance of autonomous navigation for a tracked robot. The proposed model named “layer normalization dueling double deep Q-network” has been trained in a virtual environment and then implemented it to a tracked robot for testing in a real-world scenario. Depth images have been used instead of RGB images to preserve the temporal information. Features are extracted using convolutional neural networks, and actions are derived using the dueling double deep Q-network. The input data has been normalized before each convolutional layer, which reduces the covariate shift by 69%. This end-to-end network architecture of the proposed model provides stability to the network, relieves the burden of computational cost, and converges in much less number of episodes. Compared with three Q-variant models, the proposed model demonstrates outstanding performance in terms of episodic reward and convergence rate. The proposed model took 12.8% fewer episodes for training compared to other models.

Vision-Based Autonomous Navigation Approach for a Tracked Robot Using Deep Reinforcement Learning

Tracked robots need to achieve safe autonomous steering in various changing environments. In this article, a novel end-to-end network architecture is proposed for tracked robots to learn collision-free autonomous navigation through deep reinforcement learning. Specifically, this research improved the learning time and exploratory nature of the robot by normalizing the input data and injecting parametric noise into the network parameters. Features were extracted from the four consecutive depth images by deep convolutional neural networks, which were used to derive the tracked robot. In addition, a comparison was made between three Q-variant models in terms of average reward, variance, and dispersion across episodes. Also, a detailed statistical analysis was performed to measure the reliability of all the models. The proposed model was superior in all the environments. It is worth noting that our proposed model, layer normalisation dueling double deep Q-network (LND3QN), could be directly transferred to a real robot without any fine-tuning after being trained in a simulation environment. The proposed model also demonstrated outstanding performance in several cluttered real-world environments considering both static and dynamic obstacles.

Towards simultaneous localization and mapping tolerant to sensors and software faults: Application to omnidirectional mobile robot

In the last few years, simultaneous localization and mapping became an important topic of research in the robotics community. This article proposes an approach for autonomous navigation of mobile robots in faulty situations. The main objective is to extend the fault tolerance strategy to simultaneous localization and mapping in presence of sensor faults or software faults in the data fusion process. Fault detection and isolation technique is performed based on duplication–comparison method and structured residuals. The proposed fault tolerance approach is based on the extended Kalman filter for simultaneous localization and mapping when an absolute localization sensor is available. The validation of the proposed approach and the extended Kalman filter for simultaneous localization and mapping algorithm is performed from experiments employing an omnidrive mobile robot, equipped with embedded sensors, namely: wheel encoders, gyroscope, two laser rangefinders and external sensor for the absolute position (indoor global positioning system). The obtained results demonstrate the effectiveness of the proposed approach where it was found that its fault tolerance performance is based essentially on the selected residuals and the values of the fault detection thresholds to be used for the fault detection and isolation.

NYK’s Approach for Autonomous Navigation – Structure of Action Planning System and Demonstration Experiments

NYK Line and its group companies have been conducting research projects towards autonomous ship operation. This paper shows the structure of our Action Planning System (APS), which integrates several support functions with man-machine interface, targeted at supporting decision making for seafarers. By conducting risk assessment with reference to class guidelines for autonomous ships, the safety of APS is verified. This paper also introduces our demonstration experiment of this system, which will be conducted in FY2019.