Visual Control System Research Articles

YOLO based deep learning on needle-type dashboard recognition for autopilot maneuvering system

Developing a fully automatic auxiliary flying system with robot maneuvering is feasible. This study develops a control vision system that can read all kind of needle-type meters. The vision device in this study implements a modified YOLO-based object detection model to recognize the airspeed readings from the needle-type dashboard. With this approach, meter information in the cockpit is replaced by a single camera and a powerful edge-computer for future autopilot maneuvering purpose. A modified YOLOv4-tiny model by adding the Spatial Pyramid Pooling (SPP) and the Bidirectional Feature Pyramid Network (BAFPN) to the Neck region of the convolutional neural networks (CNN) structure is implemented. The Taguchi method for acquiring a set of optimum hyperparameters for the CNN is applied. An improved deep learning network with higher Mean Average precision (mAP) compared with conventional YOLOv4-tiny and possessing a higher Frames Per Second (FPS) value than YOLOv4 is deployed successfully. Established a self-control system using a camera to receive airspeed indications from the designed virtual needle-type dashboard. Moreover, the dashboard’s pointer is controlled by applying the proposed control method, which contains PID control in addition to the pointer’s rotation angle recognition. A modified YOLOv4-tiny model with a fabricated system for visual dynamical recognition control is implemented successfully. The feasibility of bettering mean accuracy precision and frame per second in achieving autopilot maneuvering is verified.

Experimental and computational study of arc discharges in powerful alternate arc heater

The paper deals with the results of computational and experimental research on the formation process of electric arc discharges in a three-phase Zvezda-type alternating current arc heater. A flow modelling in the arc heater, as well as an experiment on providing a visual control for the location and the shape of arc discharges were conducted. A device of a visual control system is presented. The analysis of the results obtained from experimental research is carried out.

Weed Management in Ridge Tillage Systems—A Review

Although different modifications of ridge tillage (RT) systems exist in different regions around the world, the positive impacts of RT on crop yields and weed management are quite similar. This review gives a comprehensive summary of different forms of RT and highlights the benefits of RT for crop growth, mainly due to better access to soil moisture, nutrients and light. In temperate areas, RT can accelerate crop emergence because soil temperature is usually higher on the ridge. These stimulating effects increase crop competitiveness against weeds especially in the early period of crop development until canopy closure. RT with crops placed on the top of ridges can also be used for automatically guiding inter-row hoes and intra-row band sprayers. The ridges can replace automatic vision control systems for hoeing and band spraying, which are needed for precise weeding in conventional flat seedbeds. Therefore, RT can be considered a possible platform for smart/robotic weeding. This paper introduces a new RT system using real-time kinematic (RTK) global satellite navigation systems (GNSS) for the ridging and seeding of maize and soybean on top of recompacted ridges. Straight ridges with precise positioning data were used to guide mechanical weeding elements precisely along the crop rows. Simultaneously, weeds in the valleys were suppressed by living mulches. Field experiments with this new technology in maize showed 85.5% weed dry biomass suppression compared to an untreated control and a slightly higher weed control efficacy than mechanical weeding in flat seedbeds.

Recognition and Localization of Target Images for Robot Vision Navigation Control

This paper focuses on a visual navigation control system for mobile robots, recognizing target images and intelligent algorithms for the navigation system’s path tracking and localization techniques. This paper examines the recognition and localization of target images based on the visual navigation control of mobile robots. It proposes an efficient marking line method for recognizing and localization target images. Meanwhile, a fuzzy control method with smooth filtering and high efficiency is designed to improve the stability of robot operation, and the feasibility is verified in different scenarios. The corresponding image acquisition system is developed according to the characteristics of the experimental environment, and the acquired images are preprocessed to obtain corrected grayscale images. Then, target image recognition and linear fitting are performed to obtain target image positioning. The system calculates the angle and distance of the mobile robot, offsetting the target image in real time, adjusting the output signal, and controlling the mobile robot to realize path tracking. The comparison of sensor data and path tracking algorithm results during the experiment shows that the path tracking algorithm achieves good results with an angular deviation of ±1.5°. The application of RANSAC algorithm and improved Hough algorithm was analyzed in visual navigation control, and the two navigation line detection algorithms based on the image characteristics of the target image were improved in the optical detection area of the navigation line for the shortcomings of the two algorithms in visual navigation control, and the algorithms before and after the improvement were compared.

Application of Visual Centralized Control System in Underground Coal Transportation

Application of Visual Centralized Control System in Underground Coal Transportation

Intelligent Visual Quality Control System Based on Convolutional Neural Networks for Holonic Shop Floor Control of Industry 4.0 Manufacturing Systems

Intelligent Visual Quality Control System Based on Convolutional Neural Networks for Holonic Shop Floor Control of Industry 4.0 Manufacturing Systems

Visual servoing with deep reinforcement learning for rotor unmanned helicopter

Visual servoing is a key approach to achieve visual control for the rotor unmanned helicopter. The challenges of the inaccurate matrix estimation and the target loss restrict the performance of the visual servoing control systems. This work proposes a novel visual servoing controller using the deep Q-network to achieve an efficient matrix estimation. A deep Q-network learning agent learns a policy estimating the interaction matrix for visual servoing of a rotor unmanned helicopter using continuous observation. The observation includes a combination of feature errors. The current matrix and the desired matrix constitute the action space. A well-designed reward guides the deep Q-network agent to get a policy to generate a time-varying linear combination between the current matrix and the desired matrix. Then, the interaction matrix is calculated by the linear combination. The potential mapping between the observation and the interaction matrix is learned by cascading the deep neural network layers. Experimental results show that the proposed method achieves faster convergence and lower target loss probability in tracking than the visual servoing methods with the fixed parameter.

Co-Design Secure Control Based on Image Attack Detection and Data Compensation for Networked Visual Control Systems

The incomplete and untrue data caused by cyber attacks (e.g., image information leakage and tampering) will affect control performance and even lead to system instability. To address this problem, a novel co-design secure control method based on image attack detection and data compensation for networked visual control systems (NVCSs) is proposed. Firstly, the existing problems of NVCSs under image attacks are analysed, and a co-design secure control method including image encryption, watermarking-based attack detection and online data compensation is presented. Then, a detector based on double-layer detection mechanism of timeout and digital watermarking is designed for real-time, integrity and authenticity discrimination of the images. Furthermore, according to the detection results, an online compensation scheme based on cubic spline interpolation and post-prediction update is proposed to reduce the effect of cumulative errors and improve control performance. Finally, the online compensation scheme is optimized by considering the characters of networked inverted pendulum visual control systems, and experimental results demonstrate the feasibility and effectiveness of the proposed detection and control method.

IMPROVEMENT OF ROBOTIC SYSTEMS BASED ON VISUAL CONTROL

The subject of research in the article is the use of adaptive visual control in flexible integrated robotic systems. The goal of the work is the integration of visual control facilities into automated control systems for transport and handling operations of flexible integrated production. The article solves the following tasks: analyze the application of visual control methods in robotics, consider methods for improving adaptive visual work control systems, formulate the basic requirements for adaptive visual control systems, and develop a control model for a mobile robot in the space of a flexible integrated production systems and computer vision systems. To solve the set tasks, the methods of set theory, methods of automatic control theory, and methods of the theory of image processing were used. The following results were obtained: the analysis of visual control systems was carried out from the point of view of solving the problems of flexible integrated systems of modern production; the adaptive visual control scheme was improved by introducing a declarative workspace model and a functional model of a flexible integrated system; the main requirements and tasks of adaptive visual control systems are formulated; considered the main stages of processing visual information and their practical implementation, including multi-zone workspaces; a model of visual control of a mobile robot in a flexible integrated production workspace has been developed; the practical tasks of managing mobile platforms have been solved. Conclusions: the use of adaptive visual control in a production environment will allow combining the elements of flexible integrated production distributed in space, providing monitoring, control and refinement of control processes in real time, the functioning of intelligent control tools, which will improve the quality of control processes.

A software and hardware system for synchronous control of the optical system for visual control of dynamic objects

This work deals with the development of the software for the system of synchronous control of the optical system of visual control over dynamic objects and the creation of algorithms for resolving conflicts of simultaneous illumination by several light sources. The software for the microcontroller for the system hardware is created as part of the work. An application for remote configuration of the device is developed. Algorithms for time-separation of signals for switching on lighting are implemented.

Automatic assistance system for visual control of targets

The article discusses the possibility of creating a system for automating the process of searching and assessing the danger of targets using machine vision and existing ship systems. The technologies for processing images from the ship’s camera system are analyzed: object recognition by a ready-made convolutional neural retrained network, as well as object detection by SURF algorithms. The parameters of the danger of goals within the framework of the problem to be solved have been analyzed and developed, and a structural analysis has been elaborated. The functionality of the automatic operation of the system as a closed circuit has been considered. To achieve the technical result in the system as an integrated part of the ship’s equipment based on cognitive information processing, a composite information object from the ship’s sensors and control systems was introduced. It is decomposed into blocks in which the associated logical information processing takes place using basic modules consisting of adaptive cells-approximators as learning elements capable of independently processing information. The following units are included in the local area network: an information support unit (ISU), a decision making unit (DMU), a command-organizational unit (COU) with an operator’s workstation. The ISU is connected by a two-way communication output with aggregate LAN units, the other output is connected in series with a situation determination unit and the “model of the surrounding space” block. The survey was carried out on board a modern LNG gas carrier using ship systems.

Real-time vision-assisted electrochemical machining with constant inter-electrode gap

Electrochemical machining (ECM) has attracted increasing interest as an effective technique to realize high machining accuracy for difficult-to-machine conductive materials. The inter-electrode gap (IEG) has a direct effect on the ECM process performance in terms of process stability, machining accuracy, and surface quality. To understand the evolution of the IEG during the ECM process, in-situ observation and control of the IEG are required. This paper explores the use of visual assistance as a passive, in-line IEG measurement control technique to realize a consistent IEG throughout the ECM process. Here, a vision-assisted constant-IEG electrochemical machining process is proposed. The accuracy of this technique was confirmed via correspondence between the real-time IEG size and pixel data. A specially designed visual tracking and control system to implement a constant IEG in the ECM and a decision-making system to ensure process stability are presented. An experimental platform was designed and implemented, and a series of experiments were carried out on 3D structures of Inconel 718 alloy. The experimental results demonstrate that the proposed method can clearly identify IEG information from the machining region, and effectively restrict IEG fluctuation to a small value in ECM, which ultimately improved the machining accuracy and surface quality of the manufactured parts. Additionally, the machined surface presented preferential dissolution along the grain boundary and generated carbide and oxide products of Ti and Nb elements, because of enrichment of Ti and Nb elements and depletion of Cr, Fe, and Ni elements in the grain boundary regions.

System Control Scheme and Design Idea of Soft Sensing Model for Heavy Metal Ion Concentration

This paper combines ion exchange technology and soft sensor technology to design and improve a series of integrated automatic control and module distributed combination of new ion exchange system. At the same time, a visual water management and control system for industrial heavy metal wastewater treatment is developed based on the industrial IoT platform of eForceControl. Through information collection and data integration of heavy metal pollutants with different characteristics in the process flow, a variety of ion exchange materials in the distribution module are optimized. Combined with big data analysis and ion sensing, it realizes online monitoring and treatment of water pollution factors, optimizes production capacity management of industrial system, and is committed to solving problems such as low efficiency of traditional heavy metal wastewater treatment, single data analysis mode and chaotic operation and control.

Power marketing and distribution network remote visual control system based on big data architecture

As the main power transmission system connecting power supply equipment and power customers, whether the distribution network is safe and reliable in production and operation directly determines whether the work capacity of power supply equipment and power consumption equipment can be improved. The application of big data centralized control system can help us timely and accurately grasp the reasons for the decline of security and reliability of distribution network, and provide safe and stable power energy for power customers while improving the operation efficiency of power supply system. On the basis of previous research results, the author systematically analyzes the role of big data centralized control system in distribution network production and operation management, and puts forward some suggestions on applying big data centralized control system to distribution network production and operation management according to his many years of work experience. The comparative experiment proves that the fault location accuracy of the system designed in this paper is higher than 90%, which effectively increases the fault detection accuracy.

Is One Eye Better Than Two When Viewing Pictorial Art, Revisited

In paintings of a pictorial nature, the artist exploits the use of the traditional monocular cues to depth to provide a visual sense of three-dimensionality onto a two-dimensional surface. This illusory sense of depth can be enhanced using a variety of simple strategies. These will be described in the context of the underlying visual science and basic bioengineering control system concepts, with expansion to include non-pictorial renderings.

Deep learning-based visual control assistant for assembly in Industry 4.0

Product assembly is a crucial process in manufacturing plants. In Industry 4.0, the offer of mass-customized products is expanded, thereby increasing the complexity of the assembling phase. This implies that operators should pay close attention to small details, potentially resulting in errors during the manufacturing process owing to its high level of complexity. To mitigate this, we propose a novel architecture that evaluates the activities of an operator during manual assembly in a production cell so that errors in the manufacturing process can be identified, thus avoiding low quality in the final product and reducing rework and waste of raw materials or time. To perform this assessment, it is necessary to use state-of-the-art computer vision techniques, such as deep learning, so that tools, components, and actions may be identified by visual control systems. We develop a deep-learning-based visual control assembly assistant that enables real-time evaluation of the activities in the assembly process so that errors can be identified. A general-use language is developed to describe the actions in assembly processes, which can also be used independently of the proposed architecture. Finally, we generate two datasets with annotated data to be fed to the deep learning methods, the first for the recognition of tools and accessories and the second for the identification of basic actions in manufacturing processes. To validate the proposed method, a set of experiments are conducted, and high accuracy is obtained.

Wykonywanie obowiązków ustawowych pomimo braku wydanych przepisów wykonawczych w zakresie prowadzenia wizyjnego systemu miejsca magazynowania odpadów. Glosa do wyroku Wojewódzkiego Sądu Administracyjnego w Białymstoku z dnia 9 czerwca 2020 r. (II SA/Bk 16/20, CBOSA)

<p>The Voivodeship Administrative Court in Bialystok, in its judgement of 9 June 2020, dismissing the complaint of the limited liability company in H., on the post-inspection orders of the Podlaskie Voivodeship Inspector for Environmental Protection in Bialystok of 6 November 2019, considered the necessity to fulfill the obligation under Article 25 para. 6a of the Act on Waste, i.e. the operation of a visual control system for the place of waste storage, in force, in the opinion of the Court, from February 2019, despite the lack of executive acts issued by the competent minister. Determining the requirements for the visual control system of the place of storage or storage of waste, the minimum requirements for the technical devices of the vision control system and the requirements for the storage and sharing of the recorded image, guided by the need to enable supervision over the activities in the field of waste management, was established pursuant to the provisions of the Regulation of 29 August 2019 which entered into force in December 2019. In the opinion of the Court, the provisions of the Act on Waste do not have to be met together with the provisions issued on the basis of the Regulation, as the standards of the implementing acts only supplement the general conditions resulting from the statutory provisions.</p>

Control of the Test Rig for Spring Stiffness Measurement with Hydraulic Integrated Actuator and Computer Vision

The paper deals with the control of the test rig for measurement of the stiffness of the spring. The test rig is equipped with integrated hydraulic actuator and multilayer control system with computer vision to automate the testing process and to increase the safety of the operating staff during the testing. The integrated hydraulic actuator allows to control the piston velocity and position and to produce the force needed for the spring compression and stiffness measurement. It has an integrated manifold with four proportional flow control valves connected as the full hydraulic resistance bridge. Each control valve is controlled separately using the input signal. That makes it possible to realize different connections and control strategies of the hydraulic cylinder using the different control algorithms. The hydraulic circuit was modelled and simulated, and the achieved simulation results are presented in the paper. The test rig is controlled using the microprocessor control system and is it equipped by different transducers – position transducer, pressure transducers and force transducer to be able to collect data important for the stiffness characteristic evaluation. The upper level of the control system is created by the vision control system. The goal of the implementation of the machine vision is to decrease the number of the manual operations done by the technician by the spring testing and to extend the automatic testing cycle of the automatic computer vision-controlled movement of the actuator at the beginning of the testing cycle and finally to increase the safety of the operating staff. The structure and functionality of the designed control system, experimentally obtained results and evaluated stiffness characteristic of the tested spring are presented in the following chapters.

Visual Servo Control of Industrial Robot Based on Convolutional Neural Network

Aiming at the problem of poor flexibility of traditional industrial robot, the vision system and industrial robot control system are combined to propose a visual servo control method based on Faster RCNN. The image information of the target object is collected through the camera, and the Convolutional Neural Network(CNN) is used to detect the entire picture, then the category and position of the object to be grasped are obtained, which improves the control performance of the robot system. Experiments show that the visual servo robot control system can quickly complete the position recognition of different objects. Mean average precision of this method can reach 0.867. The maximum error between the recognized grasping position and the actual position on the coordinate axis does not exceed 3.4 mm, the detection time of each picture in the GPU environment is 17.1ms, which has certain guiding significance in the visual servo control of industrial robots.

Machine Vision and Radio-Frequency Identification (RFID) based Real-Time Part Traceability in a Learning Factory

Visual inspection-based quality control systems are deployed to meet the growing demand of high-quality products. In this paper, an inexpensive RFID (Radio-Frequency Identification) technology and a machine vision system have been integrated within an existing learning factory. RFID technology is used to trace a product/part from its origin, enabling the visibility of the product/part’s entire movement in the value chain. The goals of the paper are to track the workpieces in real time (i) to provide immediate feedback, through visibility, to operators and floor managers and (ii) to develop a database to trace parts, in future, backward through value chain. The system can be used to predict the probability of defective parts in the value chain using machine learning algorithms. This will help the manufacturers to trace the parts in value chain at any point of time including aftersales. Traceability has been a pain point for manufacturers during product recalls.