RESUMEN

To address the issue of automated apple harvesting in orchards, we propose a YOLOv5-RACF algorithm for identifying apples and calculating apple diameters. This algorithm employs the robot operating dystem (ROS) to control the robot's locomotion system, Lidar mapping, and navigation, as well as the robotic arm's posture and grasping operations, achieving automated apple harvesting and placement. The tests were conducted in an actual orchard environment. The algorithm model achieved an average apple detection accuracy (mAP@0.5) of 98.748% and a (mAP@0.5:0.95) of 90.02%. The time to calculate the diameter of one apple was 0.13 s, with a measurement accuracy within an error range of 1-3 mm. The robot takes an average of 9 s to pick an apple and return to the initial pose. These results demonstrate the system's efficiency and reliability in real agricultural environments.

RESUMEN

PURPOSE: Autonomous navigation of catheters and guidewires can enhance endovascular surgery safety and efficacy, reducing procedure times and operator radiation exposure. Integrating tele-operated robotics could widen access to time-sensitive emergency procedures like mechanical thrombectomy (MT). Reinforcement learning (RL) shows potential in endovascular navigation, yet its application encounters challenges without a reward signal. This study explores the viability of autonomous guidewire navigation in MT vasculature using inverse reinforcement learning (IRL) to leverage expert demonstrations. METHODS: Employing the Simulation Open Framework Architecture (SOFA), this study established a simulation-based training and evaluation environment for MT navigation. We used IRL to infer reward functions from expert behaviour when navigating a guidewire and catheter. We utilized the soft actor-critic algorithm to train models with various reward functions and compared their performance in silico. RESULTS: We demonstrated feasibility of navigation using IRL. When evaluating single- versus dual-device (i.e. guidewire versus catheter and guidewire) tracking, both methods achieved high success rates of 95% and 96%, respectively. Dual tracking, however, utilized both devices mimicking an expert. A success rate of 100% and procedure time of 22.6 s were obtained when training with a reward function obtained through 'reward shaping'. This outperformed a dense reward function (96%, 24.9 s) and an IRL-derived reward function (48%, 59.2 s). CONCLUSIONS: We have contributed to the advancement of autonomous endovascular intervention navigation, particularly MT, by effectively employing IRL based on demonstrator expertise. The results underscore the potential of using reward shaping to efficiently train models, offering a promising avenue for enhancing the accessibility and precision of MT procedures. We envisage that future research can extend our methodology to diverse anatomical structures to enhance generalizability.

Asunto(s)

RESUMEN

In the domain of mobile robot navigation, conventional path-planning algorithms typically rely on predefined rules and prior map information, which exhibit significant limitations when confronting unknown, intricate environments. With the rapid evolution of artificial intelligence technology, deep reinforcement learning (DRL) algorithms have demonstrated considerable effectiveness across various application scenarios. In this investigation, we introduce a self-exploration and navigation approach based on a deep reinforcement learning framework, aimed at resolving the navigation challenges of mobile robots in unfamiliar environments. Firstly, we fuse data from the robot's onboard lidar sensors and camera and integrate odometer readings with target coordinates to establish the instantaneous state of the decision environment. Subsequently, a deep neural network processes these composite inputs to generate motion control strategies, which are then integrated into the local planning component of the robot's navigation stack. Finally, we employ an innovative heuristic function capable of synthesizing map information and global objectives to select the optimal local navigation points, thereby guiding the robot progressively toward its global target point. In practical experiments, our methodology demonstrates superior performance compared to similar navigation methods in complex, unknown environments devoid of predefined map information.

RESUMEN

This research unveils a cutting-edge navigation system for deep space missions that utilizes cosmic microwave background (CMB) sensor readings to enhance spacecraft positioning and velocity estimation accuracy significantly. By exploiting the Doppler-shifted CMB spectrum and integrating it with optical measurements for celestial navigation, this approach employs advanced data processing through the Unscented Kalman Filter (UKF), enabling precise navigation amid the complexities of space travel. The simulation results confirm the system's exceptional precision and resilience in deep space missions, marking a significant advancement in astronautics and paving the way for future space exploration endeavors.

RESUMEN

This article describes the design and construction journey of a self-developed unmanned surface vehicle (USV). In order to increase the accessibility and lower the barrier of entry we propose a low-cost (under EUR 1000) approach to the vessel construction with great adaptability and customizability. This design prioritizes minimal power consumption as a key objective. It focuses on elucidating the intricacies of both the design and assembly processes involved in creating an economical USV. Utilizing easily accessible components, the boat outlined in this study has been already participated in the 1st Aegean Ro-boat Race 2023 competition and is tailored for entry into similar robotic competitions. Its primary functionalities encompass autonomous sea navigation coupled with sophisticated collision avoidance capabilities. Finally, we studied reinforcement learning strategies for constructing a robust intelligent controller for the task of USV navigation under disturbances and we show some preliminary simulation results we have obtained.

RESUMEN

The application of autonomous navigation technology of electric crawler tractors is an important link in the development of intelligent greenhouses. Aiming at the characteristics of enclosed and narrow space and uneven ground potholes in greenhouse planting, to improve the intelligence level of greenhouse electric crawler tractors, this paper develops a navigation system of electric crawler tractors for the greenhouse planting environment based on LiDAR technology. The navigation hardware system consists of five modules: the information perception module, the control module, the communication module, the motion module, and the power module. The software system is composed of three layers: the application layer, the data processing layer, and the execution layer. The developed navigation system uses LiDAR, Inertial Measurement Unit (IMU) and wheel speed sensor to sense the greenhouse environment and the crawler tractor's information, employs the Gmapping algorithm to build the greenhouse environment map, and utilizes the adaptive Monte Carlo positioning algorithm for positioning. The simulation test of different global path planning algorithms in Matlab shows that the A* algorithm obtains the optimal overall global path. In the scene of map 5, the path planned by the A* algorithm is the most significant, and the number of inflection points is reduced by 40.00% and 87.50%, respectively; meanwhile, the path length is the same as that of the Dijkstra algorithm, but the runtime is reduced by 68.87% and 81.49%, respectively; compared with the RRT algorithm, the path length is reduced by 7.27%. Therefore, the A* algorithm and the Dynamic Window Approach (DWA) method are used for tractor navigation and obstacle avoidance, which ensures global path optimality while also achieving effective local path planning for obstacle avoidance. The test results suggest that the maximum lateral deviation of the built map is 6 cm, and the maximum longitudinal deviation is 16 cm, which meets the requirement of map accuracy. Additionally, the results of the navigation accuracy test indicate that the maximum lateral deviation of navigation is less than 13 cm, the average lateral deviation is less than 7 cm, and the standard lateral deviation is less than 8 cm. The maximum heading deviation is less than 14°, the average heading deviation is less than 7°, and the standard deviation is less than 8°. These results show that the developed navigation system meets the navigation accuracy requirements of electric crawler tractors in the greenhouse environment.

RESUMEN

This study aims to enhance the safety and efficiency of port navigation by reducing ship collision accidents, minimizing environmental risks, and optimizing waterways to increase port throughput. Initially, a three-dimensional map of the port's waterway, including data on water depth, rocks, and obstacles, is generated through laser radar scanning. Visual perception technology is adopted to process and identify the data for environmental awareness. Single Shot MultiBox Detector (SSD) is utilized to position ships and obstacles, while point cloud data create a comprehensive three-dimensional map. In order to improve the optimal navigation approach of the Rapidly-Exploring Random Tree (RRT), an artificial potential field method is employed. Additionally, the collision prediction model utilizes K-Means clustering to enhance the Faster R-CNN algorithm for predicting the paths of other ships and obstacles. The results indicate that the RRT enhanced by the artificial potential field method reduces the average path length (from 500 to 430 m), average time consumption (from 30 to 22 s), and maximum collision risk (from 15 to 8%). Moreover, the accuracy, recall rate, and F1 score of the K-Means + Faster R-CNN collision prediction model reach 92%, 88%, and 90%, respectively, outperforming other models. Overall, these findings underscore the substantial advantages of the proposed enhanced algorithm in autonomous navigation and collision prediction in port waterways.

RESUMEN

Nowadays, trajectory control is a significant issue for unmanned micro aerial vehicles (MAVs) due to large disturbances such as wind and storms. Trajectory control is typically implemented using a proportional-integral-derivative (PID) controller. In order to achieve high accuracy in trajectory tracking, it is essential to set the PID gain parameters to optimum values. For this reason, separate gain values are set for roll, pitch and yaw movements before autonomous flight in quadrotor systems. Traditionally, this adjustment is performed manually or automatically in autotune mode. Given the constraints of narrow orchard corridors, the use of manual or autotune mode is neither practical nor effective, as the quadrotor system has to fly in narrow apple orchard corridors covered with hail nets. These reasons require the development of an innovative solution specific to quadrotor vehicles designed for constrained areas such as apple orchards. This paper recognizes the need for effective trajectory control in quadrotors and proposes a novel neural network-based approach to tuning the optimal PID control parameters. This new approach not only improves trajectory control efficiency but also addresses the unique challenges posed by environments with constrained locational characteristics. Flight simulations using the proposed neural network models have demonstrated successful trajectory tracking performance and highlighted the superiority of the feed-forward back propagation network (FFBPN), especially in latitude tracking within 7.52745 × 10-5 RMSE trajectory error. Simulation results support the high performance of the proposed approach for the development of automatic flight capabilities in challenging environments.

RESUMEN

The integration of machine learning and robotics brings promising potential to tackle the application challenges of mobile robot navigation in industries. The real-world environment is highly dynamic and unpredictable, with increasing necessities for efficiency and safety. This demands a multi-faceted approach that combines advanced sensing, robust obstacle detection, and avoidance mechanisms for an effective robot navigation experience. While hybrid methods with default robot operating system (ROS) navigation stack have demonstrated significant results, their performance in real time and highly dynamic environments remains a challenge. These environments are characterized by continuously changing conditions, which can impact the precision of obstacle detection systems and efficient avoidance control decision-making processes. In response to these challenges, this paper presents a novel solution that combines a rapidly exploring random tree (RRT)-integrated ROS navigation stack and a pre-trained YOLOv7 object detection model to enhance the capability of the developed work on the NAV-YOLO system. The proposed approach leveraged the high accuracy of YOLOv7 obstacle detection and the efficient path-planning capabilities of RRT and dynamic windows approach (DWA) to improve the navigation performance of mobile robots in real-world complex and dynamically changing settings. Extensive simulation and real-world robot platform experiments were conducted to evaluate the efficiency of the proposed solution. The result demonstrated a high-level obstacle avoidance capability, ensuring the safety and efficiency of mobile robot navigation operations in aviation environments.

RESUMEN

This survey reviews advances in 3D object detection approaches for autonomous driving. A brief introduction to 2D object detection is first discussed and drawbacks of the existing methodologies are identified for highly dynamic environments. Subsequently, this paper reviews the state-of-the-art 3D object detection techniques that utilizes monocular and stereo vision for reliable detection in urban settings. Based on depth inference basis, learning schemes, and internal representation, this work presents a method taxonomy of three classes: model-based and geometrically constrained approaches, end-to-end learning methodologies, and hybrid methods. There is highlighted segment for current trend of multi-view detectors as end-to-end methods due to their boosted robustness. Detectors from the last two kinds were specially selected to exploit the autonomous driving context in terms of geometry, scene content and instances distribution. To prove the effectiveness of each method, 3D object detection datasets for autonomous vehicles are described with their unique features, e. g., varying weather conditions, multi-modality, multi camera perspective and their respective metrics associated to different difficulty categories. In addition, we included multi-modal visual datasets, i. e., V2X that may tackle the problems of single-view occlusion. Finally, the current research trends in object detection are summarized, followed by a discussion on possible scope for future research in this domain.

RESUMEN

One of the important issues being explored in Industry 4.0 is collaborative mobile robots. This collaboration requires precise navigation systems, especially indoor navigation systems where GNSS (Global Navigation Satellite System) cannot be used. To enable the precise localization of robots, different variations of navigation systems are being developed, mainly based on trilateration and triangulation methods. Triangulation systems are distinguished by the fact that they allow for the precise determination of an object's orientation, which is important for mobile robots. An important feature of positioning systems is the frequency of position updates based on measurements. For most systems, it is 10-20 Hz. In our work, we propose a high-speed 50 Hz positioning system based on the triangulation method with infrared transmitters and receivers. In addition, our system is completely static, i.e., it has no moving/rotating measurement sensors, which makes it more resistant to disturbances (caused by vibrations, wear and tear of components, etc.). In this paper, we describe the principle of the system as well as its design. Finally, we present tests of the built system, which show a beacon bearing accuracy of Δφ = 0.51°, which corresponds to a positioning accuracy of ΔR = 6.55 cm, with a position update frequency of fupdate = 50 Hz.

RESUMEN

Endoscopes navigate within the human body to observe anatomical structures with minimal invasiveness. A major shortcoming of their use is their narrow field-of-view during navigation in large, hollow anatomical regions. Mosaics of endoscopic images can provide surgeons with a map of the tool's environment. This would facilitate procedures, improve their efficiency, and potentially generate better patient outcomes. The emergence of magnetically steered endoscopes opens the way to safer procedures and creates an opportunity to provide robotic assistance both in the generation of the mosaic map and in navigation within this map. This paper proposes methods to autonomously navigate magnetic endoscopes to 1) generate endoscopic image mosaics and 2) use these mosaics as user interfaces to navigate throughout the explored area. These are the first strategies, which allow autonomous magnetic navigation in large, hollow organs during minimally invasive surgeries. The feasibility of these methods is demonstrated experimentally both in vitro and ex vivo in the context of the treatment of twin-to-twin transfusion syndrome. This minimally invasive procedure is performed in utero and necessitates coagulating shared vessels of twin fetuses on the placenta. A mosaic of the vasculature in combination with autonomous navigation has the potential to significantly facilitate this challenging surgery.

Asunto(s)

RESUMEN

Robotic missions for solar farm inspection demand agile and precise object detection strategies. This paper introduces an innovative keypoint-based object detection framework specifically designed for real-time solar farm inspections with UAVs. Moving away from conventional bounding box or segmentation methods, our technique focuses on detecting the vertices of solar panels, which provides a richer granularity than traditional approaches. Drawing inspiration from CenterNet, our architecture is optimized for embedded platforms like the NVIDIA AGX Jetson Orin, achieving close to 60 FPS at a resolution of 1024 ×1376 pixels, thus outperforming the camera's operational frequency. Such a real-time capability is essential for efficient robotic operations in time-critical industrial asset inspection environments. The design of our model emphasizes reduced computational demand, positioning it as a practical solution for real-world deployment. Additionally, the integration of active learning strategies promises a considerable reduction in annotation efforts and strengthens the model's operational feasibility. In summary, our research emphasizes the advantages of keypoint-based object detection, offering a practical and effective approach for real-time solar farm inspections with UAVs.

RESUMEN

The challenge of precise dynamic positioning for mobile robots is addressed through the development of a multi-inertial navigation system (M-INSs). The inherent cumulative sensor errors prevalent in traditional single inertial navigation systems (INSs) under dynamic conditions are mitigated by a novel algorithm, integrating multiple INS units in a predefined planar configuration, utilizing fixed distances between the units as invariant constraints. An extended Kalman filter (EKF) is employed to significantly enhance the positioning accuracy. Dynamic experimental validation of the proposed 3INS EKF algorithm reveals a marked improvement over individual INS units, with the positioning errors reduced and stability increased, resulting in an average accuracy enhancement rate exceeding 60%. This advancement is particularly critical for mobile robot applications that demand high precision, such as autonomous driving and disaster search and rescue. The findings from this study not only demonstrate the potential of M-INSs to improve dynamic positioning accuracy but also to provide a new research direction for future advancements in robotic navigation systems.

RESUMEN

Robust evidence suggests that humans explore their environment using a combination of topological landmarks and coarse-grained path integration. This approach relies on identifiable environmental features (topological landmarks) in tandem with estimations of distance and direction (coarse-grained path integration) to construct cognitive maps of the surroundings. This cognitive map is believed to exhibit a hierarchical structure, allowing efficient planning when solving complex navigation tasks. Inspired by human behaviour, this paper presents a scalable hierarchical active inference model for autonomous navigation, exploration, and goal-oriented behaviour. The model uses visual observation and motion perception to combine curiosity-driven exploration with goal-oriented behaviour. Motion is planned using different levels of reasoning, i.e., from context to place to motion. This allows for efficient navigation in new spaces and rapid progress toward a target. By incorporating these human navigational strategies and their hierarchical representation of the environment, this model proposes a new solution for autonomous navigation and exploration. The approach is validated through simulations in a mini-grid environment.

RESUMEN

Microrobotics has opened new horizons for various applications, especially in medicine. However, it also witnessed challenges in achieving maximum optimal performance. One key challenge is the intelligent, autonomous, and precise navigation control of microrobots in fluid environments. The intelligence and autonomy in microrobot control, without the need for prior knowledge of the entire system, can offer significant opportunities in scenarios where their models are unavailable. In this study, two control systems based on model-free deep reinforcement learning were implemented to control the movement of a disk-shaped magnetic microrobot in a real-world environment. The training and results of an off-policy SAC algorithm and an on-policy TRPO algorithm revealed that the microrobot successfully learned the optimal path to reach random target positions. During training, the TRPO exhibited a higher sample efficiency and greater stability. The TRPO and SAC showed 100% and 97.5% success rates in reaching the targets in the evaluation phase, respectively. These findings offer basic insights into achieving intelligent and autonomous navigation control for microrobots to advance their capabilities for various applications.

RESUMEN

At the beginning of a project or research that involves the issue of autonomous navigation of mobile robots, a decision must be made about working with traditional control algorithms or algorithms based on artificial intelligence. This decision is not usually easy, as the computational capacity of the robot, the availability of information through its sensory systems and the characteristics of the environment must be taken into consideration. For this reason, this work focuses on a review of different autonomous-navigation algorithms applied to mobile robots, from which the most suitable ones have been identified for the cases in which the robot must navigate in dynamic environments. Based on the identified algorithms, a comparison of these traditional and DRL-based algorithms was made, using a robotic platform to evaluate their performance, identify their advantages and disadvantages and provide a recommendation for their use, according to the development requirements of the robot. The algorithms selected were DWA, TEB, CADRL and SAC, and the results show that-according to the application and the robot's characteristics-it is recommended to use each of them, based on different conditions.

RESUMEN

The relative position of the orchard robot to the rows of fruit trees is an important parameter for achieving autonomous navigation. The current methods for estimating the position parameters between rows of orchard robots obtain low parameter accuracy. To address this problem, this paper proposes a machine vision-based method for detecting the relative position of orchard robots and fruit tree rows. First, the fruit tree trunk is identified based on the improved YOLOv4 model; second, the camera coordinates of the tree trunk are calculated using the principle of binocular camera triangulation, and the ground projection coordinates of the tree trunk are obtained through coordinate conversion; finally, the midpoints of the projection coordinates of different sides are combined, the navigation path is obtained by linear fitting with the least squares method, and the position parameters of the orchard robot are obtained through calculation. The experimental results show that the average accuracy and average recall rate of the improved YOLOv4 model for fruit tree trunk detection are 5.92% and 7.91% higher, respectively, than those of the original YOLOv4 model. The average errors of heading angle and lateral deviation estimates obtained based on the method in this paper are 0.57° and 0.02 m. The method can accurately calculate heading angle and lateral deviation values at different positions between rows and provide a reference for the autonomous visual navigation of orchard robots.

RESUMEN

This paper presents a dataset of bird's eye chilies in a single farm for semantic segmentation. The dataset is generated using two cameras that are aligned left and right forming a stereo-vision video capture. By analyzing the disparity between corresponding points in the left and right images, algorithms can calculate the relative distance of objects in the scene. This depth information is useful in various applications, including 3D reconstruction, object tracking, and autonomous navigation. The dataset consists of 1150 left and right compressed images extracted from ten sets of stereo videos taken at ten different locations within the chili farm from the same ages of the bird's eye chilies. Since the dataset is used for semantic segmentation, the ground truth images of manually semantic segmented images are also provided in the dataset. The dataset can be used for 2D and 3D semantic segmentation of the bird's eye view chili farm. Some of the object classes in this dataset are the sky, living things, plantation, flat, construction, nature, and misc.

RESUMEN

Mobile robots are playing an increasingly significant role in social life and industrial production, such as searching and rescuing robots, autonomous exploration of sweeping robots, and so on. Improving the accuracy of autonomous navigation of mobile robots is a hot issue to be solved. However, traditional navigation methods are unable to realize crash-free navigation in an environment with dynamic obstacles, more and more scholars are gradually using autonomous navigation based on deep reinforcement learning (DRL) to replace overly conservative traditional methods. But on the other hand, DRL's training time is too long, and the lack of long-term memory easily leads the robot to a dead end, which makes its application in the actual scene more difficult. To shorten training time and prevent mobile robots from getting stuck and spinning around, we design a new robot autonomous navigation framework which combines the traditional global planning and the local planning based on DRL. Therefore, the entire navigation process can be transformed into first using traditional navigation algorithms to find the global path, then searching for several high-value landmarks on the global path, and then using the DRL algorithm to move the mobile robot toward the designated landmarks to complete the final navigation, which makes the robot training difficulty greatly reduced. Furthermore, in order to improve the lack of long-term memory in deep reinforcement learning, we design a feature extraction network containing memory modules to preserve the long-term dependence of input features. Through comparing our methods with traditional navigation methods and reinforcement learning based on end-to-end depth navigation methods, it shows that while the number of dynamic obstacles is large and obstacles are rapidly moving, our proposed method is, on average, 20% better than the second ranked method in navigation efficiency (navigation time and navigation paths' length), 34% better than the second ranked method in safety (collision times), 26.6% higher than the second ranked method in success rate, and shows strong robustness.