Mobile Robot Platform Research Articles

Design and Development of a Mobile Robotic System for Aircraft Wing Fuel Tank Inspection

<div class="section abstract"><div class="htmlview paragraph">This paper presents the design concept behind a novel remote visual inspection robotic system for fighter jet aircraft wing fuel tank inspection. This work is part of a larger research project which focuses on design, simulation, physical prototyping and experimental validation of a robotic system. Whereas this paper specifically focuses on the development concept of locomotion design choice for the robot. Therefore without an effective mobility method the robot will not be able to fulfill its purpose to access the hazardous confined spaces of the fuel tank. Aircraft wing fuel tank inspection is a challenging area of maintenance which requires a considerable amount of preparation and involvement of several tasks in order to conduct effective Visual and Non Destructive Inspection. The environment of an aircraft wing fuel tank poses several challenges due to both physical and atmospheric constraints which can be detrimental to human personal. This paper introduces an effective locomotion design which should allow the robot to enter and maneuver within confined spaces. The robot is relatively small, approximately 70mm in height and width yet, flexible enough to move within the restricted spaces of the wing. The mobile robot platform is a combination of small track systems that articulate like a snake. An additional mobile platform deploys an inspection sensor to reach the spaces that are unreachable by the robot body. Like other proposed robotic systems this particular proposal differs as it allows the robot to enter from the root of the wing and reach the narrower spaces towards with the wing tip. This paper highlights the stakeholder requirements to illustrate the foundation of the robotic system design. An overview of current complications of wing fuel tank inspection is presented and the analysis of current proposed robotic systems for wing fuel tank inspection. An engineering design methodology approach is followed for this project. Several locomotion methods are evaluated and an innovative locomotion method is illustrated with the use of CAD models. The desired outcome of this research is to eliminate the entry or close contact with the fuel tank by human personal.</div></div>

Autonomous Mobile Robot Implemented in LEGO EV3 Integrated with Raspberry Pi to Use Android-Based Vision Control Algorithms for Human-Machine Interaction

Robotic applications, such as educational programs, are well-known. Nonetheless, there are challenges to be implemented in other settings, e.g., mine detection, agriculture support, and tasks for industry 4.0. The main challenge consists of robotic operations supported by autonomous decision using sensed-based features extraction. A prototype of a robot assembled using mechanical parts of a LEGO MINDSTORMS Robotic Kit EV3 and a Raspberry Pi controlled through servo algorithms of 2D and 2D1/2 vision approaches was implemented to tackle this challenge. This design is supported by simulations based on image, position, and a hybrid scheme for visual servo controllers. Practical implementation is operated using navigation guided by running up image-based visual servo control algorithms embedded in a Raspberry Pi that uses a control criterion based on error evolution to compute the difference between a target and sensed image. Images are collected by a camera installed on a mobile robotic platform manually and automatically operated and controlled using the Raspberry Pi. An Android application to watch the images by video streaming is shown here, using a smartphone and a video related to the implemented robot’s operation. This kind of robot might be used to complete field reactive tasks in the settings mentioned above, since the detection and control approaches allow self-contained guidance.

AN INTEGRATED AVOIDING AND APPROACHING HUMAN SYSTEMS FOR MOBILE SERVICE ROBOTS IN DYNAMIC SOCIAL ENVIRONMENTS

In this study, we present a socially aware navigation framework for mobile service robots by integrating the systems proposed in our previous studies into a completed mobile robot navigation system, including the human detection and tracking system that is used to detect and track humans; the encoder and laser-based localization and mapping system that is used to determine the position of the robot in the environment; the approaching pose estimation system that is used to estimate the approaching pose of the robot to the human; and the social timed elastic band-based motion planning system that is used to socially navigate the mobile robots to approach the human. In addition, in the paper we describe in detail the motor control system and the design of our mobile robot platform, which is then utilized to conduct experiments in the real-world environments. We verify the feasibility and usefulness of the proposed socially aware robot navigation framework through a series of experiments in a corridor-like environment. The experimental results show that our proposed framework is able to drive the mobile robots to both avoid and approach the humans, providing the safety and comfort for the humans and socially acceptable behaviors for the mobile service robots in the dynamic social environments.

Intelligent intra-row robotic weeding system combining deep learning technology with a targeted weeding mode

Mechanical weed control is the most promising weed control method for weed management in organic agriculture. Among the various available methods, inter-row weeding technology is relatively mature, and researchers have focused on intra-row weeding technology. The objective of this study was to develop a new intelligent intra-row mechanical robotic weeding system based on deep learning for crop and weed detection. The system consists of a mobile robot platform and two intelligent weeding units. The mobile robot platform provides power support as well as operating conditions for the weeding units. A targeted weeding pattern was proposed based on the deep learning detection results, and protected and targeted weeding zones were established by strict criteria to reduce crop injury rates. In addition, three weed control knives were designed according to different field environments. Field trials showed that among the three weed knives, the plough-surface weeding knife was the most effective and most suitable knife for flat cultivation, and the wedge weeding knife was most suitable for ridge planting. Based on the obtained results, the best weeding method was determined, and the experiment was repeated using a wedge weeding knife under ridge planting conditions. The final weed removal rate was 85.91%, and the crop injury rate was 1.17%. The results demonstrated the feasibility of the proposed intra-row weed control method.

Development of dual-arm mobile robot platform based on ROS

Background: With the increasing demand of mobile robots in warehousing, logistics and service fields, simple planar motion is difficult to meet the task requirements of complex environment. The combination of mobile robot and cooperative robot is helpful to improve the dexterity of robot movement and expand the application of robots. Methods: Aiming at the application requirements of dual-arm robots and mobile robots in practical applications, this paper designed the hardware of a platform, built a simulation platform based on ROS (Robot Operating System), and designed the actual software control framework. Finally, the feasibility of the platform design was verified by the coupling motion experiment of the two robots. Results: We have established a simulation of the dual-arm mobile platform in ROS, designed the actual software control framework, and verified the feasibility of the platform design through experiments. Conclusions: The mobile platform can meet a variety of application requirements and lay the foundation for subsequent development.

Estimation of the Energy Consumption of an All-Terrain Mobile Manipulator for Operations in Steep Vineyards

A heterogeneous robotic system that can perform various tasks in the steep vineyards of the Mediterranean region was developed and tested as part of the HEKTOR—Heterogeneous Autonomous Robotic System in Viticulture and Mariculture—project. This article describes the design of hardware and an easy-to-use method for evaluating the energy consumption of the system, as well as, indirectly, its deployment readiness level. The heterogeneous robotic system itself consisted of a flying robot—a light autonomous aerial robot (LAAR)—and a ground robot—an all-terrain mobile manipulator (ATMM), composed of an all-terrain mobile robot (ATMR) platform and a seven-degree-of-freedom (DoF) torque-controlled robotic arm. A formal approach to describe the topology and parameters of selected vineyards is presented. It is shown how Google Earth data can be used to make an initial estimation of energy consumption for a selected vineyard. On this basis, estimates of energy consumption were made for the tasks of protective spraying and bud rubbing. The experiments were conducted in two different vineyards, one with a moderate slope and the other with a much steeper slope, to evaluate the proposed estimation method.

Cross-modal distillation for RGB-depth person re-identification

Person re-identification is a key challenge for surveillance across multiple sensors. Prompted by the advent of powerful deep learning models for visual recognition, and inexpensive RGB-D cameras and sensor-rich mobile robotic platforms, e.g. self-driving vehicles, we investigate the relatively unexplored problem of cross-modal re-identification of persons between RGB (color) and depth images. The considerable divergence in data distributions across different sensor modalities introduces additional challenges to the typical difficulties like distinct viewpoints, occlusions, and pose and illumination variation. While some work has investigated re-identification across RGB and infrared, we take inspiration from successes in transfer learning from RGB to depth in object detection tasks. Our main contribution is a novel method for cross-modal distillation for robust person re-identification, which learns a shared feature representation space of person’s appearance in both RGB and depth images. In addition, we propose a cross-modal attention mechanism where the gating signal from one modality can dynamically activate the most discriminant CNN filters of the other modality. The proposed distillation method is compared to conventional and deep learning approaches proposed for other cross-domain re-identification tasks. Results obtained on the public BIWI and RobotPKU datasets indicate that the proposed method can significantly outperform the state-of-the-art approaches by up to 16.1% in mean Average Precision (mAP), demonstrating the benefit of the distillation paradigm. The experimental results also indicate that using cross-modal attention allows to improve recognition accuracy considerably with respect to the proposed distillation method and relevant state-of-the-art approaches.11Code: https://github.com/frhf/cross-modal-distillation-reidentification.

Review and application of Edge AI solutions for mobile collaborative robotic platforms

Edge AI technologies have emerged as a powerful solution for gathering and processing data in real-time, supporting the development of advanced applications for process monitoring, planning and control. Edge AI devices coupled with different sensory systems can be used for facilitating the synergetic human-robot collaboration at the shop floor level. This paper reviews edge AI devices and solutions that may potentially support the operation of cells or workstations comprising human operators and mobile robotic platforms. The use of edge AI-capable devices for operator tracking in real-world applications and the potential of edge AI technologies towards developing and deploying process digital twins are discussed. The main goal is to investigate how edge AI solutions may improve the performance of processes where a collaborative mobile manipulator works alongside a human operator in a safe and reliable manner. Lastly, the use of advanced communication frameworks, such as the Robot Operating System (ROS), is discussed in the context of supporting the seamless interaction between a collaborative mobile platform and Edge AI devices.

On Reducing the Localisation Error of Modern Mobile Robotic Platforms

From drug delivery in hospitals to transportation of goods in warehouses, mobile robotic platforms have a wide variety of applications in industry today. The integration of mobile platforms with a manipulator opens up the possibility of more practical applications. However, this also introduces a new problem regarding the localisation accuracy of the platform, which in turn affects the accuracy of the manipulation. This paper details a comparative study of different strategies for improving the accuracy of pick and place tasks using a mobile robotic platform with an integrated arm. These strategies include fine positioning, LiDAR feedback and vision system feedback.

A Cost-Efficient Hybrid UHF RFID and Odometry-Based Mobile Robot Self-Localization Technique With Centimeter Precision

This paper presents an indoor localization approach that determines the absolute position of a mobile robot platform with centimeter precision by fusing RFID localization results based on cost-effective, standard passive UHF RFID technology with the robot’s odometry data. The mobile robot platform is equipped with a multistatic UHF RFID interrogation system, and several RFID tags are arbitrarily placed within the localization environment, serving as landmarks. The RFID localization concept is based on phase evaluations. To overcome the problem of ambiguous position estimations due to the 2π-phase ambiguity of the RFID signal phase and mitigate the linearization problem of the nonlinear system, a novel algorithm based on an iterative multihypothesis Kalman filter is introduced. A realistic simulation setup is developed to validate the proposed filter algorithm. By tracking a UHF RFID-equipped mobile robot platform in a real-world office environment, the proposed approach is also practically tested in terms of real-time capability, everyday suitability, and multipath resistance. Given that centimeter precision is only achieved in environments with weak multipath propagation, the RFID localization results are fused with odometry data provided by the robot. This effectively compensates for offset and drift in the odometry sensor, achieving a root-mean-square localization error of 2.7 cm.

MODELS AND TOOLS FOR DEBUGGING AND TESTING MOBILE SYSTEMS FOR NEURO-LIKE CRYPTOGRAPHIC PROTECTION OF DATA TRANSMISSION

The work revealed the need for providing cryptographic protection and immunity to data transmission and control commands when using the mobile robotic platform as well as the importance of taking into account the limitations regarding dimensions, energy consumption and productivity. It was found that one of the ways to meet the requirements of cryptographic protection is the use of neuro-like networks. Their feature is the ability to pre-calculate the weight coefficients that will be used when encrypting/decrypting data. It is suggested that during neuro-like encryption/decryption of data, the key should be generated taking into account the architecture of the neuro-like network (the number of neurons, the number of inputs and their bit rate), the matrix of weight coefficients and the table for masking. It was determined that a neural network with pre-calculated weight coefficients makes it possible to use a table-algorithmic method for data encryption/decryption, which is based on the operations of reading from memory, adding and shifting. Limitations regarding dimensions, energy consumption and performance are analyzed. They can be overcome during implementation by using a universal processor core supplemented with specialized FPGA hardware for neuro-like elements. That is the combined use of software and specialized hardware ensures the effective implementation of neuro-like data encryption/decryption algorithms and management teams. Models and tools for debugging and testing a neuro-like cryptographic system are presented. A model of the preliminary settings of the neuro-like data encryption system has been developed, the main components of which are the former of the neuro-like network architecture, the calculator of weight coefficient matrices and the calculator of tables of macro-partial products. A model of the process of neuro-like encryption of control commands using a table-algorithmic method has been developed. Models for testing and debugging blocks of encryption (decryption), encoding (decoding), and masking (unmasking) of data have been developed, which, due to the use of reference values for comparison, ensure an increase in the quality of testing and debugging of the cryptographic system. A cryptographic system was developed, which, as a result of a dynamic change in the type of neuro-like network architecture and the values of weighting coefficients, mask codes and barker-like code, provides an increase in the crypto-resistance of data transmission. Testing of the simulation model was carried out on the example of message transmission for various configurations of a cryptographic system.

Rao-Blackwellized Particle Filter with Neural Network using Low-Cost Range Sensor in Indoor Environment

Implementation of Rao-Blackwellized Particle Fil-ter (RBPF) in grid-based simultaneous localization and mapping (SLAM) algorithm with range sensors is commonly developed by using sensor with dense measurements such as laser rangefinder. In this paper, a more cost convenient solution was explored where implementation of array of infrared sensors equipped on a mobile robot platform was used. The observation from array of infrared sensors are noisy and sparse. This adds more uncertainty in the implementation of SLAM algorithm. To compensate for the high uncertainties from robot’s observations, neural network was integrated with the grid-based SLAM algorithm. The result shows that the grid-based SLAM algorithm with neural network has better accuracy compared to the grid-based SLAM algorithm without neural network for the aforementioned mobile robot implementation. The algorithm improves the map accuracy by 21% and reduce robot’s state estimate error significantly. The better performance is due to the improvement in accuracy of grid cells’ occupancy value. This affects the importance weight computation in RBPF algorithm hence resulting a better map accuracy and robots state estimate. This finding shows that a promising grid-based SLAM algorithm can be obtained by using merely array of infrared sensors as robot’s observation.

A Map Construction Method Based on the Cognitive Mechanism of Rat Brain Hippocampus

The entorhinal-hippocampus structure in the mammalian brain is the core area for realizing spatial cognition. However, the visual perception and loop detection methods in the current biomimetic robot navigation model still rely on traditional visual SLAM schemes and lack the process of exploring and applying biological visual methods. Based on this, we propose a map construction method that mimics the entorhinal-hippocampal cognitive mechanism of the rat brain according to the response of entorhinal cortex neurons to eye saccades in recent related studies. That is, when mammals are free to watch the scene, the entorhinal cortex neurons will encode the saccade position of the eyeball to realize the episodic memory function. The characteristics of this model are as follows: 1) A scene memory algorithm that relies on visual saccade vectors is constructed to imitate the biological brain's memory of environmental situation information matches the current scene information with the memory; 2) According to the information transmission mechanism formed by the hippocampus and the activation theory of spatial cells, a localization model based on the grid cells of the entorhinal cortex and the place cells of the hippocampus was constructed; 3) Finally, the scene memory algorithm is used to correct the errors of the positioning model and complete the process of constructing the cognitive map. The model was subjected to simulation experiments on publicly available datasets and physical experiments using a mobile robot platform to verify the environmental adaptability and robustness of the algorithm. The algorithm will provide a basis for further research into bionic robot navigation.

A Novel Lidar-Assisted Monocular Visual SLAM Framework for Mobile Robots in Outdoor Environments

In this paper, a novel 3D lidar-assisted monocular visual SLAM (LAMV-SLAM) framework is proposed for mobile robots in outdoor environments. LAMV-SLAM can run in real-time without a GPU and build a dense map with real scale. An online photometric calibration thread is integrated into LAMV-SLAM to eliminate the photometric disturbances in images. The tracking thread combines the lidar and vision data to estimate and refine the frame-to-frame transformation. In this thread, the depth fusion algorithm is proposed to provide accurate depth values for the extracted visual features by combining the lidar points, and a novel two-stage optimization method is proposed to utilize the fused lidar-vision data to estimate the camera transformation with the real scale. A parallel mapping thread generates new map points based on depth-filter and lidar-vision data fusion. A loop closing thread further reduces the accumulative errors of the system. To verify the accuracy and efficiency of the system, we evaluated the proposed pipeline on the KITTI odometry benchmark, and our LAMV-SLAM achieves a 0.81% of relative position drift while running at over 3X real-time speed. To verify the robustness of the system in challenging environments, experiments were carried out on the NCLT and NuScenes datasets. Moreover, real-world experiments were applied to our mobile robot platform to show the practicability and validity of the proposed approach.

A MODEL FOR IMPROVING THE STRENGTH CHARACTERISTICS OF THE ELECTROMECHANICAL DRIVE OF A MOBILE ROBOT

Mobile robots are increasingly used in the most diverse spheres of human activities; accordingly, it is essential to ensure their reliable functioning, which in turn determines efficiency. Using appropriate calculations during design, it is possible to increase reliability and reduce the metal consumption of the machine samples being created. It is crucial that such calculations consider the loading modes in which the vehicle is used. The purpose of the presented work is to increase the technical and operational indicators of the electromechanical drive of mobile robots by selecting the input parameters in combination with the appropriate methods and techniques of design and mathematical modelling. In order to achieve the specified goal, the following main tasks of the research are defined: firstly, to improve the model of increasing reliability and reducing the metal consumption of mechanical components of mobile robots; and secondly, to calculate the mechanical components of mobile robots using the proposed model. Providing the necessary margin of strength with a simultaneous reduction in metal density is necessary for improving the electromechanical drive of a mobile robot and improving its characteristics in general. The paper presents a model and developed an algorithm for increasing the reliability and reducing the metal consumption of mechanical components of mobile robots. The method includes geometric, kinematic, dynamic, energy, technical and economic indicators' calculations, as well as strength and stiffness calculations. The calculations were performed for a small mobile robot with an electromechanical transmission, and the results of a study of the reliability and strength characteristics of the shaft of the mobile robotics platform were presented. The case of turning a mobile robot with the realization of the maximum torque, which is transmitted to one of the tracks, is considered. Based on the kinematic scheme of the electric transmission, a solid-state model of one of its elements (the traction star shaft of the crawler motor) was developed, for which, based on the schematized Serensen – Kinasoshvili diagram, the margin of safety was determined. The proposed model has been examined and successfully used to construct the experimental samples of mobile robots.

Design and simulation of a mobile robot platform for navigation and obstacle detection

Mobile platforms are expected to gain access to risk zones and hazardous environment. A typical example is the infectious disease environment to deliver items to a sick patient. The robot is aimed to manoeuver round flat grounds in indoor environment. Computer aided design (CAD) models of the selected concepts were developed in Fusion360 and imported into SolidWorks to optimize and improve the design. The design is focused on the development of the wheelbase. Arduino Microcontroller was the system and codes control board and it was developed using the Arduino software. The motor driver was used to drive the DC motor for robot navigation with ultrasonic sensor for obstacle detection at a range of 20 cm. Result shows that the robot was able to navigate round flat ground while detecting obstacles within 20 cm.

An Open-Source Low-Cost Mobile Robot System With an RGB-D Camera and Efficient Real-Time Navigation Algorithm

Currently, mobile robots are developing rapidly and are finding numerous applications in the industry. However, several problems remain related to their practical use, such as the need for expensive hardware and high power consumption levels. In this study, we build a low-cost indoor mobile robot platform that does not include a LiDAR or a GPU. Then, we design an autonomous navigation architecture that guarantees real-time performance on our platform with an RGB-D camera and a low-end off-the-shelf single board computer. The overall system includes SLAM, global path planning, ground segmentation, and motion planning. The proposed ground segmentation approach extracts a traversability map from raw depth images for the safe driving of low-body mobile robots. We apply both rule-based and learning-based navigation policies using the traversability map. Running sensor data processing and other autonomous driving components simultaneously, our navigation policies perform rapidly at a refresh rate of 18 Hz for control command, whereas other systems have slower refresh rates. Our methods show better performances than current state-of-the-art navigation approaches within limited computation resources as shown in 3D simulation tests. In addition, we demonstrate the applicability of our mobile robot system through successful autonomous driving in an indoor environment.

Abstraction in teaching ways of control engineering to support the understanding of mathematics behind Industry 4.0 – a Hungarian approach

Modern technologies, as well as the technologies that will emerge in the near future, pose a great challenge to researchers and educators in education and educational methodology. Teaching the intuitive use of mathematical tools, developing the ability to abstract, using appropriate theoretical and practical models and demonstration tools are all required to equip engineering students with the right mindset and practice to meet the challenges of the modern era. In this paper, we present the theoretical and practical tools that are used in daily practice in the context of control engineering to help engineering students develop their mindset. First, the development of an appropriate mathematical toolbox and mindset is introduced, followed by the development of abstraction skills during the teaching process. Interdisciplinary, mechanical and robotic models and tools are used to educate students about the complexity of engineering. The paper will start from mechanics, then present an interdisciplinary example in the field of blood glucose control, and, in addition, highly nonlinear demonstration tools will be presented, and a mobile robot platform for education, developed in collaboration with engineering students, will be presented, which is suitable for teaching in the fields of sensor fusion, artificial intelligence and others.

Integrated Mobile Robotic Platform Model

Introduction. The “Smart Agroˮ committee of Research and Education Center “Engineering of the Future” has identified a number of tasks relevant for improving the efficiency of precision, soil-protecting and conservation agriculture. One of these tasks is the development of a digital multi-agent system, which provides a number of services for agricultural enterprises, developers and manufacturers of agricultural machinery. The purpose of the present study is to model an autonomous mobile robotic platform, including the development of software and hardware for trajectory control. Materials and Methods. To solve the problem, there are used modern CAx systems and their applications, the methods of 3D and full-body modeling, and the method of numerical solution of problems in solid mechanics. To expand and improve the standard functionality of CAx-systems (SolidWorks) in the software implementation of trajectory control algorithms, the methods and technologies of programming using API SolidWorks, VisualStudio C++ (MFC, ATL, COM) are used, and to build physical full-scale models ‒ Arduino and fischertechnik platforms. Results. The result of the study is a software and hardware module of trajectory control for an integrated (physical and virtual) model of a mobile robotic platform, which can be provided to the consumer as a service for technology autonomation. For the developed integrated model, control algorithms for various types of trajectories were tested. Discussion and Conclusion. The developed integrated software and hardware model of trajectory control can be used by developers and manufacturers of agricultural machinery, and directly by agro-enterprises for implementing typical technological processes. A feature of the implementation is an open hardware and software interface that provides the integration of mobile robotic platforms based on a digital multi-agent system.

Constructing Floor Plan through Smoke Using Ultra Wideband Radar

Floor plan construction has been one of the key techniques in many important applications such as indoor navigation, location-based services, and emergency rescue. Existing floor plan construction methods require expensive dedicated hardware (e.g., Lidar or depth camera), and may not work in low-visibility environments (e.g., smoke, fog or dust). In this paper, we develop a low-cost Ultra Wideband (UWB)-based system (named UWBMap) that is mounted on a mobile robot platform to construct floor plan through smoke. UWBMap leverages on low-cost and off-the-shelf UWB radar, and it is able to construct an indoor map with an accuracy comparable to Lidar (i.e., the state-of-the-art). The underpinning technique is to take advantage of the mobility of radar to form virtual antennas and gather spatial information of a target. UWBMap also eliminates both robot motion noise and environmental noise to enhance weak reflection from small objects for the robust construction process. In addition, we overcome the limited view of single radar by combining multi-view from multiple radars. Extensive experiments in different indoor environments show that UWBMap achieves a map construction with a median error of 11 cm and a 90-percentile error of 26 cm, and it operates effectively in indoor scenarios with glass wall and dense smoke.