Description:TECHNICAL FIELD
[0001] The present disclosure relates, in general, to artificial intelligence-guided robotics, and more specifically, relates to an automated system and method for bowser inspection at the cement plant.

BACKGROUND
[0002] Various technologies are currently employed for bowser inspection including visual inspection, ultrasonic testing and radiography. Visual inspection is the most common form of bowser inspection carried out by the industry and is completely manual. This involves a manual visual inspection of the bowser's external and internal components. These inspections are normally aided by tools such as flashlights, mirrors, or cameras to access hard-to-reach areas.
[0003] The visual inspection of bowser has limitations as follows:
• Limited Accessibility: Some areas of the bowser may be difficult or impossible to access visually, especially internal components that are not easily visible.
• Human Error: Visual inspection relies on the inspector's ability to identify defects or issues, which can be subject to human error. Inspectors may miss defects due to fatigue, distraction, or lack of experience.
• Subjectivity: Visual inspection is subjective, and different inspectors may have different interpretations of what constitutes a defect or issue. This can lead to inconsistent inspection results.
• Time-Consuming: Visual inspection can be a time-consuming process, especially if the bowser is large or has complex internal components that require disassembly.
• Safety Concerns: Visual inspections are hazardous as the inspector needs to climb the bowser to conduct the inspection additionally, he may also need to access confined spaces, which may contain residual harmful chemicals. Hence manual visual inspection process has a higher risk of falls, slips, ingestion of harmful chemicals and other accidents.
[0004] The ultrasonic testing technique uses high-frequency sound waves to detect internal defects or cracks in the bowser's structure. The sound waves are transmitted through the material, and the reflected waves are analyzed to determine the location and size of any defects.
[0005] The ultrasonic testing has limitations as follows:
• Material Limitations: Ultrasonic testing is limited to certain materials, such as metals and plastics. Materials with high attenuation or low acoustic impedance may not be suitable for ultrasonic testing.
• Surface Condition: Ultrasonic testing requires a smooth and flat surface for accurate results. Rough or irregular surfaces can produce false readings or make it difficult to detect defects.
• Calibration: Ultrasonic testing requires regular calibration to ensure accurate results. The calibration process can be time-consuming and can impact inspection efficiency.
• Operator Skill: Ultrasonic testing requires a skilled operator to perform the inspection and interpret the results accurately. The accuracy of the results depends on the skill and experience of the operator.
[0006] The radiography involves the use of X-rays or gamma rays to penetrate the bowser's material and produce an image that can be analyzed for defects or cracks. Radiography is commonly used for detecting defects in welded or jointed areas of the bowser. The radiography-based bowser inspection has limitations as follows:
• Radiation Exposure: Radiography involves the use of ionizing radiation, which is hazardous to the health of the inspector and other personnel in the area. Strict safety measures must be followed to minimize the risk of radiation exposure.
• Operator skill: Radiography produces images that require interpretation by a trained inspector. Interpretation is subjective and may vary depending on the experience and skill of the inspector. Radiography equipment requires specialized training to operate which cannot be performed by normal factory operators.
• Surface Condition: Radiography requires a smooth and flat surface for accurate imaging. Rough or irregular surfaces can produce false readings or make it difficult to detect defects. Most bowser has irregular surfaces due to residual material/operational wear and tear.
• Thickness Limitations: Radiography has limitations in terms of the thickness of the material being inspected. Thicker materials may require higher levels of radiation exposure or specialized equipment to produce accurate images.
[0007] Therefore, it is desired to overcome the drawbacks, shortcomings, and limitations associated with existing solutions by providing the integration of AI and robotics that holds the potential to overcome many of the limitations in existing bowser inspection technologies.

OBJECTS OF THE PRESENT DISCLOSURE
[0008] An object of the present disclosure relates, to artificial intelligence-guided robotics, and more specifically, relates to an automated system and method for bowser inspection at the cement plant.
[0009] Another object of the present disclosure is to provide an automated inspection system that significantly reduces the risk associated with hazardous materials or confined spaces by eliminating the need for human inspectors to enter the bowser. This safety improvement ensures that workers are not exposed to potentially dangerous environments during the inspection process.
[0010] Another object of the present disclosure is to provide an automated inspection system that detects the material and enables accurate data collection from various angles within the bowser's inlet hole using AI and robotics that are designed to work with rough irregular surfaces and complex locations.
[0011] Another object of the present disclosure is to provide an automated inspection system that works with a variety of materials, not just metals and plastics, this can improve the versatility of the inspection system and enable it to detect defects in a wider range of materials.
[0012] Another object of the present disclosure is to provide an automated inspection system with AI algorithms that are trained to interpret the images produced by the inspection system, reducing the subjectivity of image interpretation and improving the accuracy of defect detection.
[0013] Another object of the present disclosure is to provide an automated inspection system that reduces the need for labor-intensive and time-consuming manual processes. The system's efficiency leads to cost savings by optimizing resource utilization, minimizing operational downtime, and reducing the chances of costly errors or oversights.
[0014] Another object of the present disclosure is to provide an automated inspection system that detects the presence or absence of materials within the bowser. These insights provide actionable information for maintenance and operational decisions, ensuring timely interventions when necessary.
[0015] Another object of the present disclosure is to provide an automated inspection system that reduces the need for skilled operators by automating the inspection process. This can improve the consistency and reliability of inspections, regardless of the operator's experience level.
[0016] Another object of the present disclosure is to provide an automated inspection system that can be easily scaled to accommodate various sizes and types of bowsers across different industries. This adaptability makes the solution versatile and applicable to a wide range of scenarios.
[0017] Another object of the present disclosure is to provide an automated inspection system that generates detailed and standardized reports based on the collected data. These reports provide clear documentation of the inspection results, which can be valuable for compliance, auditing, and quality assurance purposes.
[0018] Yet another object of the present disclosure is to provide the robotic arm that is equipped with proximity sensors that have been programmed to serve as a safety feature for the robotic arm. These sensors are designed to detect the presence of objects in the vicinity of the robotic arm and send signals to the control system, which can then initiate appropriate actions to prevent collisions or other hazards. This technology enhances the safety and reliability of the robotic arm in various applications.

SUMMARY
[0019] The present disclosure relates to artificial intelligence-guided robotics, and more specifically, relates to an automated system and method for bowser inspection at the cement plant. The main objective of the present disclosure is to overcome the drawback, limitations, and shortcomings of the existing system and solution, by incorporating a four-axis robotic arm that can move inside the inlet hole of the bowser to collect data using a machine vision camera. The collected images are processed using machine learning algorithms to detect the presence or absence of material inside the bowser. By automating the inspection process using AI and robotics, significant benefits to the customer are provided, including improved safety, enhanced accuracy, increased efficiency, reduced downtime, and cost savings. Automating the inspection process with robotics and AI technology improves safety by eliminating the need for human inspectors to enter the bowser, which is hazardous due to the presence of materials or the confined space.
[0020] The present disclosure includes a jib crane adapted to be mounted on a platform for bowser inspection, the jib crane having two degrees of freedom sliding motion both in vertical and horizontal directions. The motorized hoist mounted on the jib crane serves as a trolley for x-axis movement, the motorized hoist being equipped with a guider roller for smooth sliding movement. The robotic arm is coupled to the motorized hoist, the robotic arm equipped with a first set of sensors and a second set of sensors and adapted to move inside an inlet hole of the bowser to capture a set of images of the internal surface of the bowser. The alert generation unit generates an alert for automated material detection in the bowser inspection.
[0021] The PLC integrated with the AI server operatively coupled to the robotic arm and the alert generation unit, the PLC integrated with the AI server configured to operate, upon receiving a set of data, from the first set of sensors, the robotic arm to align with the calculated position of the bowser hole. The set of data pertaining to comprehensive visual coverage and position of the browser hole. The captured set of images received from the second set of sensors is analysed to perform image annotation by identifying class labels for the detected objects within the captured images. The set of images pertains to the internal surface of the bowser, where the analysed set of images is sorted to detect the presence and absence of the material within the bowser. An alert is emitted in response to signals received from the PLC indicating the presence of the material within the bowser.
[0022] The PLC integrated with the AI server facilitates the training of deep learning models for material detection. The first set of sensors can be one or more IP cameras and second set of sensors can be machine vision camera. The one or more IP cameras integrated with the control system of the robotic arm are strategically positioned at distinct angles to provide the set of data pertaining to comprehensive visual coverage and the position of the browser hole. The PLC is configured to process the visual data captured by the one or more IP cameras and calculate the precise position of the bowser hole, wherein the PLC processes the received position information and generates control signals to adjust the positioning of the robotic arm, ensuring alignment with the detected bowser hole.
[0023] In addition, the alert generation unit includes an alarm hooter that is activated to emit alerts by the PLC upon receiving signals indicating the material presence in the bowser. The alarm hooter is configured to generate a red signal indication raised by the PLC when the material is detected within the bowser during the inspection and generate a green signal indication raised by the PLC when the bowser is inspected and deemed to be free of material presence. The alert generation unit can include communication channels for generating alerts through electronic communication i.e., email/ SMS, where the alerts are transmitted via the communication channels based on signals received from the AI server through the PLC.
[0024] Various objects, features, aspects, and advantages of the inventive subject matter will become more apparent from the following detailed description of preferred embodiments, along with the accompanying drawing figures in which like numerals represent like components.

BRIEF DESCRIPTION OF THE DRAWINGS
[0025] The following drawings form part of the present specification and are included to further illustrate aspects of the present disclosure. The disclosure may be better understood by reference to the drawings in combination with the detailed description of the specific embodiments presented herein.
[0026] FIG. 1A to 1E illustrates exemplary functional components of an automated inspection system, in accordance with an embodiment of the present disclosure.
[0027] FIG. 2 illustrates exemplary functional components of the proposed system in accordance with an embodiment of the present disclosure.
[0028] FIG. 3A to 3B illustrates exemplary integration and assembly of an automated inspection system, in accordance with an embodiment of the present disclosure.
[0029] FIG. 3C illustrates a wiring diagram of the automated inspection system, in accordance with an embodiment of the present disclosure.
[0030] FIG. 4 illustrates an exemplary flow chart of a method of bowser inspection using an automated inspection system, in accordance with an embodiment of the present disclosure.

DETAILED DESCRIPTION
[0031] The following is a detailed description of embodiments of the disclosure depicted in the accompanying drawings. The embodiments are in such detail as to clearly communicate the disclosure. If the specification states a component or feature “may”, “can”, “could”, or “might” be included or have a characteristic, that particular component or feature is not required to be included or have the characteristic.
[0032] As used in the description herein and throughout the claims that follow, the meaning of “a,” “an,” and “the” includes plural reference unless the context clearly dictates otherwise. Also, as used in the description herein, the meaning of “in” includes “in” and “on” unless the context clearly dictates otherwise.
[0033] The present disclosure relates, to artificial intelligence-guided robotics, and more specifically, relates to an automated system and method for bowser inspection at the cement plant. In industries like cement production, the bowser plays a crucial role in transporting and dispensing liquids, such as raw materials like ash and chemicals. However, the manual inspection of these essential components can be both time-consuming and perilous. To overcome this challenge, an automated inspection system that harnesses advanced robotics and artificial intelligence techniques is disclosed.
[0034] The present disclosure incorporates a four-axis robotic arm that can move inside the inlet hole of the bowser to collect data using a machine vision camera. The collected images are processed using machine learning algorithms to detect the presence or absence of material inside the bowser. By automating the inspection process using AI and robotics, significant benefits to the customer are provided, including improved safety, enhanced accuracy, increased efficiency, reduced downtime, and cost savings. Automating the inspection process with robotics and AI technology improves safety by eliminating the need for human inspectors to enter the bowser, which is hazardous due to the presence of materials or the confined space. The present disclosure can be described in enabling detail in the following examples, which may represent more than one embodiment of the present disclosure.
[0035] The advantages achieved by the system of the present disclosure can be clear from the embodiments provided herein. The integration of robotics and AI into the automated inspection system for bowser inspections brings forth enhanced safety, accurate data collection, efficiency, reduced downtime, cost savings, data-driven insights, consistency, scalability, quality documentation, and a competitive edge. This transformative approach improves safety by eliminating human exposure to hazardous environments, while also streamlining operations, reducing costs, and elevating overall inspection quality and effectiveness. The description of terms and features related to the present disclosure shall be clear from the embodiments that are illustrated and described; however, the invention is not limited to these embodiments only. Numerous modifications, changes, variations, substitutions, and equivalents of the embodiments are possible within the scope of the present disclosure. Additionally, the invention can include other embodiments that are within the scope of the claims but are not described in detail with respect to the following description.
[0036] FIG. 1A to 1E illustrate exemplary functional components of an automated inspection system, in accordance with an embodiment of the present disclosure.
[0037] Referring to FIG. 1A to 1E, automated inspection system 100 (also referred to as system 100, herein) for bowser inspection at the cement plant is disclosed. The automated inspection system 100 is implemented on the bowser that is designed for storing and transporting liquids, especially fuels, water, ash, chemicals or other liquids. The bowser is commonly used in industries such as agriculture, construction, aviation, and transportation to facilitate the transfer and distribution of fluids. The specific use and design of the bowser can vary depending on the industry and the type of liquid being transported or stored. The system 100 can include a jib structure 102, hoist 104, robotic arm 106, programmable logic controller (PLC) 110, artificial intelligence (AI) server 112, a first set of sensors 114 -1, second set of sensors 114-2 and an alert generation unit 116. The first set of sensors 114-1 can include one or more IP cameras and second set of sensors 114 can include machine vision camera. To conduct a visual inspection, the inlet hole 108 is present at the top of the bowser to access the internal bowser.
[0038] In an embodiment, FIG. 1A depicts that the jib structure 102 (also referred to as jib crane 102, herein) serves as a pivotal structure enabling precise positioning for the inspection process. The inspection of the internal bowser is performed manually by an operator using a platform that has been specifically designed for this purpose. The existing platform may be repurposed to mount the jib crane 102, which has two degrees of freedom sliding motion, vertical and horizontal motion. The jib crane 102 has the capability to carry a weight of up to 100kg, and its total length is 13.01 meters. The length of the jib crane 102 varies based on the length of the bowser and platform.0.75 HP induction motor has been utilized for sliding movement.
[0039] FIG. 1B refers to the hoist 104 which is a motorized trolley that is used for the x-axis and is controlled automatically. The length of the motorized trolley is 3290 mm and has the capability to move along two axes. It is powered by a 0.5HP induction motor, which is controlled by a control panel. This hoist 104 has been installed with a guider roller for its smooth sliding movement. To ensure the safety of the motorized trolley, a limit switch stopper has been integrated into the control system. The stopper is integrated to automatically stop the movement of the trolley when necessary.
[0040] FIG. 1C depicts the robotic arm 106 that is integrated with the machine vision camera to capture high-quality images in dusty environments. The camera is equipped with a self-lens cleaning mechanism to maintain its effectiveness. The robotic arm 106 has the capability to rotate 360 degrees along a single axis. Additionally, machine vision lights have been installed within the enclosure of the camera. The brake stepper motor, which has a torque of 26 kg/cm, is coupled with the camera enclosure. Both the stepper motor and the lights are controlled by a control panel, and the machine vision camera is connected to the AI server using a CAT 6A patch cord.
[0041] The machine vision camera enclosures are protective housings that are designed to house machine vision cameras, which are used for automated inspection, it is designed in a way that the camera field of view is not disturbed. Machine vision camera enclosures are manufactured from mild steel (MS) material, which provides high resistance to heat, shock, and vibration. Enclosures are designed to protect against harsh environmental conditions, such as dust, moisture and the like. The self-lens cleaning mechanism is installed in the camera enclosure to get quality images in a dust environment.
[0042] The machine vision lights are specialized lighting systems that are designed to provide illumination for machine vision systems to keep quality during day and night the high-quality machine vision light is installed on the enclosure of this machine vision light. The capacity of the light is 300 watts and it operated at the current 24V 3.5Amp. To supply power direct current (DC) plug and play connector are used.
[0043] Further, the robotic arm 106 is equipped with proximity sensors that have been programmed to serve as a safety feature for the robotic arm 106. These sensors are designed to detect the presence of objects in the vicinity of the robotic arm and send signals to the control system, which can then initiate appropriate actions to prevent collisions or other hazards. This technology enhances the safety and reliability of the robotic arm in various applications.
[0044] The AI server 112 enclosure is configured to keep the server in operation. The AI server is designed for the outdoor environment. To maintain the temperature, air cooling is also provided to cool down the enclosure temperature. The AI server 112 is made with MS material also it has an anti-rust layer on the surface to prevent rust occur due to changes in environmental conditions. The server enclosures are designed to protect electronic equipment from harsh conditions, such as dust, dirt, moisture, extreme temperatures, and physical impact. Industrial server enclosures are often equipped with locks, to prevent unauthorized access and theft. These enclosures typically have cable management systems that organize cables, reducing clutter and the risk of damage. Enclosures are usually made of heavy-duty materials such as MS. This server enclosure has Ingress Protection (IP55) for protection against environmental hazards.
[0045] To prevent damage caused by erroneous parking of the bowser that can cause damage to the robotic arm 106 during the inspection, the automatic bowser hole 108 detections are implemented shown in FIG. 1D. One or more IP cameras have been installed above the bowser. The one or more IP cameras as presented in the example can be three IP cameras that have been installed above the bowser to detect the position of the hole. The machine vision algorithm can calculate its position and it is integrated with the robotic arm control system. The dimension wise position information is sent to the PLC 110, which sends signals to the robotic arm 106 to move accordingly. By integrating the camera system with the robotic arm control system through PLC, the position of the hole is accurately detected and the robotic arm is moved accordingly, preventing damage caused by erroneous parking.
[0046] In an implementation of an embodiment, shown in FIG. 1E, the jib crane 102 adapted to be mounted on a platform for bowser inspection, the jib crane having two degrees of freedom sliding motion both in vertical and horizontal direction. The motorized hoist 104 mounted on the jib crane 102 serves as a trolley for x-axis movement, the motorized hoist being equipped with a guider roller for smooth sliding movement. The robotic arm 106 coupled to the motorized hoist, the robotic arm equipped with a set of sensors 114 and adapted to move inside an inlet hole of the bowser to capture a set of images of the internal surface of the bowser. The alert generation unit 116 generates an alert for automated material detection in the bowser inspection.
[0047] The PLC 110 integrated with the AI server 112 operatively coupled to the robotic arm 106 and the alert generation unit 116, the PLC integrated with the AI server configured to operate, upon receiving a set of data, from the set of sensors 114, the robotic arm 106 to align with the calculated position of the bowser hole. The set of data pertaining to comprehensive visual coverage and position of the browser hole. The captured set of images received from the set of sensors is analysed to perform image annotation by identifying class labels for the detected objects within the captured images. The set of images pertains to the internal surface of the bowser, where the analysed set of images is sorted to detect the presence and absence of the material within the bowser. The alert is emitted in response to signals received from the PLC indicating the presence of the material within the bowser.
[0048] The PLC integrated with AI server facilitates the training of deep learning models for material detection. The set of sensors can include machine vision camera 114-2 and one or more IP cameras 114-1. The one or more IP cameras integrated with control system of the robotic arm are strategically positioned at distinct angles to provide the set of data pertaining to comprehensive visual coverage and position of the browser hole. The PLC is configured to process the visual data captured by the one or more IP cameras and calculate the precise position of the bowser hole, where the PLC processes the received position information and generates control signals to adjust the positioning of the robotic arm, ensuring alignment with the detected bowser hole.
[0049] In addition, the alert generation unit 116 includes an alarm hooter that is activated to emit alerts by the PLC upon receiving signals indicating material presence in the bowser. The alarm hooter is configured to generate a red signal indication raised by the PLC when material is detected within the bowser during inspection and generate a green signal indication raised by the PLC when the bowser is inspected and deemed to be free of material presence. The alert generation unit 116 can include communication channels for generating alerts through electronic communication i.e., email/ SMS, wherein the alerts are transmitted via the communication channels based on signals received from the AI server through the PLC.
[0050] In another implementation of an embodiment, the AI for bowser internal inspection can be performed as follows:
• Capturing the images: The robotic arm 106 is equipped with the machine vision camera and machine vision light. Machine vision light is used to maintain the image quality during day and night. The machine vision camera captures the images of the internal surface of bowser and sends them to the AI-server 112. The camera is connected to AI server by using CAT 6A patch cord.
• Data annotation: In order to detect the presence/absence of material in the bowser, it is necessary to perform manual annotation using a software. The process of manual annotation involves the marking and labeling of the components in the images to facilitate their detection.
• Following are the steps for image annotation.
Step1 - Identifying and specifying class labels for the objects intended to be detected within images.
Step2 - Applying a bounding box around each instance of the specified objects in the images, thereby indicating their respective locations for detection purposes. The objects specified for detection include categories such as "extra material" and "bottom panel ok."
Step3 - Exporting the annotations in JSON format, forming a structured dataset suitable for training deep learning models for object detection.
In the case of bowser inspection, a dataset of at least 500 images is required for the detection of single material through manual annotation.
• Image sorting: In this step, the annotated images are again cleaned/checked before starting the training for AI model generation.
[0051] In an exemplary embodiment, Convolutional Neural Network (CNN) architecture is used to perform the bowser inspection, using open-source libraries and python programming code. The optimal number of iterations and checkpoints is required to effectively train and learn from the available dataset. Also, this iteration values are depended upon size and complexity of the dataset.
[0052] Once the architecture has been generated, the model is trained using the training dataset with high performance Graphics Processing Unit (GPU). The training period is depended upon the datasets and number of labels. Following are the graphics processing unit specifications. i. Gigabyte H410M-H V3 Motherboard, ii. RAM: 16 GB DDR4 Desktop Ram iii. Graphics card: NVIDIA Galaxy 3060 iv. Hard disk: 1 TB Seagate SATA Disk.
[0053] In another exemplary embodiment, the AI server system may have the following software installed.
i. Training may perform on the Ubuntu 18.04 operating system with CUDA 10.0, cu DNN v7.6.5 software also Py-Torch and OpenCV frameworks are installed for image processing.
ii. Convolutional neural network (CNN) based on the Mask-RCNN architecture is used for model training.
[0054] After the model has been trained, it is evaluated on a separate validation dataset to determine its accuracy and performance. If the model is not performing well, the architecture is adjusted or the model is retrained.
[0055] The customize software solution has been designed by understanding the problem statements. The software solution has a backend AI algorithm that selects pre-built models for real time detection to ensure accurate results. The software is integrated with an IP camera, machine vision camera, control system, and AI server. The software provides accurate results and is capable of integrating with various sub-systems. By developing a customized software solution with backend AI algorithms and integration with various systems, accurate results are obtained according to the customer's requirements.
[0056] The user-friendly dashboard is developed with the following function
• Bowser OK/NOT_OK
• Data analytics.
• Daily, monthly, and yearly report generation.
[0057] Further, the system 100 is equipped with the alarm hooter controlled by the PLC integrated with the AI server once the internal material is detected. The system may send a signal to the PLC system. The Transmission Control Protocol/Internet Protocol (TCP-IP) interface is used to communicate. The PLC raises the red signal when the material is present in the bowser, and the green signal is raised when the bowser is inspected and found to be okay. Alerts can also be generated through email and SMS communication channels. By implementing the system with the hooter, PLC, and other communication interface, alerts are generated for the presence of materials in the bowser.
[0058] Thus, the present invention overcomes the drawbacks, shortcomings, and limitations associated with existing solutions, and provides the integration of robotics and AI into the automated inspection system for bowser inspections bringing forth enhanced safety, accurate data collection, efficiency, reduced downtime, cost savings, data-driven insights, consistency, scalability, quality documentation, and a competitive edge. The ability to customize the system for different third-party application or cloud storage is provided. This transformative approach improves safety by eliminating human exposure to hazardous environments, while also streamlining operations, reducing costs, and elevating overall inspection quality and effectiveness.
[0059] FIG. 2 illustrates exemplary functional components 200 of the proposed system in accordance with an embodiment of the present disclosure.
[0060] In this aspect, system 112 may also be implemented in a variety of computing systems, such as a laptop computer, a desktop computer, a notebook, a workstation, a server, a network server, and the like. The system 112 may comprise one or more processor(s) 202. The one or more processor(s) 202 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, logic circuitries, and/or any devices that manipulate data based on operational instructions. Among other capabilities, the one or more processor(s) 202 are configured to fetch and execute computer-readable instructions stored in a memory 204 of the system 112. The memory 204 may store one or more computer-readable instructions or routines, which may be fetched and executed to create or share the data units over a network service. The memory 204 may comprise any non-transitory storage device including, for example, volatile memory such as RAM, or non-volatile memory such as EPROM, flash memory, and the like.
[0061] The system 112 may also comprise an interface(s) 206. The interface(s) 206 may comprise a variety of interfaces, for example, interfaces for data input and output devices, referred to as I/O devices, storage devices, and the like. The interface(s) 206 may facilitate communication of system 112. The interface(s) 206 may also provide a communication pathway for one or more components of the system 112. Examples of such components include, but are not limited to, processing engine(s) 208 and database 210.
[0062] The processing engine(s) 208 may be implemented as a combination of hardware and programming (for example, programmable instructions) to implement one or more functionalities of the processing engine(s) 208. In the examples described herein, such combinations of hardware and programming may be implemented in several different ways. For example, the programming for the processing engine(s) 208 may be processor-executable instructions stored on a non-transitory machine-readable storage medium and the hardware for the processing engine(s) 208 may comprise a processing resource (for example, one or more processors), to execute such instructions. In the present examples, the machine-readable storage medium may store instructions that, when executed by the processing resource, implement the processing engine(s) 208. In such examples, the system may comprise the machine-readable storage medium storing the instructions and the processing resource to execute the instructions, or the machine-readable storage medium may be separate but accessible to system 112 and the processing resource. In other examples, the processing engine(s) 208 may be implemented by electronic circuitry.
[0063] The processing engine(s) 208 may include a deep learning engine 212 and other engine(s) 214. Other engine(s) 214 can supplement the functionalities of the processing engine 208 or the system. The database 210 may comprise data that is either stored or generated as a result of functionalities implemented by any of the components of the processing engine(s) 208 or the system.
[0064] FIG. 3A to 3B illustrates exemplary integration and assembly of automated inspection system, in accordance with an embodiment of the present disclosure.
[0065] The hardware parameters and software parameters for AI based bowser inspection is shown in table 1 and table 2 below.
Capacity 100kg
Location Outdoor
Bay Length 13.01 MTR
Ambient Temp 45oC
Power Supply 415V,3 PHASE
Control Volt 110V
Crane Control Push Button
Speed HOIST C.T L.T
3 mtr / min 10 mtr / min 10mtr / min
Motor MAKE REMI REMI REMI
RATED HP CDF 40% S4 DUTY 0.5 H.P. 6 POLE 0.5 H.P. 6 POLE 0.75 H.P. 6 POLE
Brake ELECTROMAGNETIC DISKS BRAKES
Camera Movements THROUGH GEARBOX / MOTOR & WHEEL
Controller Programmable Logic Controller
Communication TCP IP Interface
AI server I7, GPU 3060,1TB- HDD,32GB RAM.
Machine Vision Light INPUT-24VOLT, 2AMP.
Machine Vision Camera 2048 X 1536
Camera Interface GigE (POE)
Monitor 19 Inch
Table 1: Hardware parameters for AI based bowser inspection

Operating System Ubuntu
Programming Language Python
Result Time 1 Sec
User Interface User Interface with OK/NOT_OK Bowser, Data Analytics, Daily Report with download function
Table 2: Software parameters for AI based bowser inspection

[0066] FIG. 3C illustrates wiring diagram of automated inspection system, in accordance with an embodiment of the present disclosure. The wiring diagram for the automated inspection system depicts the integration of multiple components to create a comprehensive inspection solution. The diagram outlines the electrical connections and interactions among key elements to ensure seamless operation.
[0067] FIG. 4 illustrates an exemplary flow chart of method of bowser inspection using an automated inspection system, in accordance with an embodiment of the present disclosure
[0068] Referring to FIG. 4, the method 400 includes at block 402, mounting the jib crane on a platform for bowser inspection, the jib crane having two degrees of freedom sliding motion both in vertical and horizontal direction. At block 404, serving the motorized hoist 104 mounted on the jib crane 102 as a trolley for x-axis movement, the motorized hoist being equipped with a guider roller for smooth sliding movement.
[0069] At block 406, operating, by the programmable logic controller (PLC) 110 integrated with an artificial intelligence (AI) server, upon receiving a set of data from the first set of sensors, a robotic arm to align with the calculated position of the bowser hole. The set of data pertaining to comprehensive visual coverage and position of the browser hole, where the robotic arm (106) coupled to the motorized hoist and equipped with the first set of sensors and second set of sensors adapted to move inside an inlet hole of the bowser to capture the set of images of the internal surface of the bowser.
[0070] At block 408, analysing, by the PLC integrated with an AI server, the captured set of images received from the second set of sensors to perform image annotation by identifying class labels for the detected objects within the captured images, the set of images pertains to the internal surface of the bowser, where the analysed set of images is sorted to detect the presence and absence of the material within the bowser. At block 410, emitting an alert in response to signals received from the PLC indicating the presence of the material within the bowser, where an alert generation unit coupled to the PLC generates the alert for automated material detection in the bowser inspection.
[0071] It will be apparent to those skilled in the art that the system 100 of the disclosure may be provided using some or all of the mentioned features and components without departing from the scope of the present disclosure. While various embodiments of the present disclosure have been illustrated and described herein, it will be clear that the disclosure is not limited to these embodiments only. Numerous modifications, changes, variations, substitutions, and equivalents will be apparent to those skilled in the art, without departing from the spirit and scope of the disclosure, as described in the claims.

ADVANTAGES OF THE PRESENT INVENTION
[0072] The present invention provides an automated inspection system that significantly reduces the risk associated with hazardous materials or confined spaces by eliminating the need for human inspectors to enter the bowser. This safety improvement ensures that workers are not exposed to potentially dangerous environments during the inspection process.
[0073] The present invention provides an automated inspection system that enables accurate data collection from various angles within the bowser's inlet hole using the robotic arm's precision movements. This accuracy ensures that the inspection process captures comprehensive and reliable data, minimizing the chances of missing potential issues.
[0074] The present invention provides an automated inspection system that enables a swift and efficient inspection process compared to manual methods. The robotic arm's controlled and systematic movements allow for consistent data collection and analysis, leading to faster inspections and quicker decision-making.
[0075] The present invention provides an automated inspection system with automation, inspections can be carried out during planned maintenance or operational downtimes, minimizing disruptions to production processes. This proactive approach helps prevent unexpected downtime caused by inspection-related delays.
[0076] The present invention provides an automated inspection system that reduces the need for labor-intensive and time-consuming manual processes. The system's efficiency leads to cost savings by optimizing resource utilization, minimizing operational downtime, and reducing the chances of costly errors or oversights.
[0077] The present invention provides an automated inspection system that detects the presence or absence of materials within the bowser. These insights provide actionable information for maintenance and operational decisions, ensuring timely interventions when necessary.
[0078] The present invention provides an automated inspection system that ensures inspections are performed consistently, eliminating variations that can occur with manual inspections. This reliability in data collection contributes to more accurate trend analysis and informed decision-making.
[0079] The present invention provides an automated inspection system that can be easily scaled to accommodate various sizes and types of bowsers across different industries. This adaptability makes the solution versatile and applicable to a wide range of scenarios.
[0080] The present invention provides an automated inspection system that generates detailed and standardized reports based on the collected data. These reports provide clear documentation of the inspection results, which can be valuable for compliance, auditing, and quality assurance purposes.
, Claims:1. An automated inspection system (100) for bowser inspection, the system comprising:
a jib crane (102) adapted to be mounted on a platform for bowser inspection, the jib crane having two degrees of freedom sliding motion both in vertical and horizontal direction;
a motorized hoist (104) mounted on the jib crane (102) serving as a trolley for x-axis movement, the motorized hoist being equipped with a guider roller for smooth sliding movement;
a robotic arm (106) coupled to the motorized hoist, the robotic arm equipped with a first set of sensors and second set of sensors and adapted to move inside an inlet hole of the bowser to capture a set of images of the internal surface of the bowser;
an alert generation unit (116) to generate an alert for automated material detection in the bowser inspection; and
a programmable logic controller (PLC) (110) integrated with an artificial intelligence (AI) server (112) operatively coupled to the robotic arm and the alert generation unit, the PLC integrated with AI server configured to:
operate, upon receiving a set of data, from the first set of sensors, the robotic arm to align with the calculated position of the bowser hole, the set of data pertaining to comprehensive visual coverage and position of the browser hole;
analyse the captured set of images received from the second set of sensors to perform image annotation by identifying class labels for the detected objects within the captured images, the set of images pertains to the internal surface of the bowser, wherein the analysed set of images is sorted to detect the presence and absence of the material within the bowser; and
emit an alert in response to signals received from the PLC indicating the presence of the material within the bowser.

2. The system as claimed in claim 1, wherein the PLC integrated with AI server facilitate the training of deep learning models for material detection.

3. The system as claimed in claim 1, wherein the first set of sensors is one or more internet protocol (IP) cameras and second set of sensors is machine vision camera.

4. The system as claimed in claim 3, wherein the one or more IP cameras integrated with control system of the robotic arm are strategically positioned at distinct angles to provide the set of data pertaining to comprehensive visual coverage and position of the browser hole.

5. The system as claimed in claim 4, wherein the PLC configured to process the visual data captured by the one or more IP cameras and calculate the precise position of the bowser hole, wherein the PLC processes the received position information and generates control signals to adjust the positioning of the robotic arm, ensuring alignment with the detected bowser hole.

6. The system as claimed in claim 1, wherein the motorized hoist comprises a limit switch stopper integrated into a control system of the motorized hoist (104) to automatically halt trolley movement when required for safety.

7. The system as claimed in claim 1, wherein the alert generation unit (116) comprises
an alarm hooter that is activated to emit alerts by the PLC upon receiving signals indicating the material presence in the bowser.

8. The system as claimed in claim 1, wherein the alarm hooter configured to:
generate a red signal indication raised by the PLC when the material is detected within the bowser during inspection; and
generate a green signal indication raised by the PLC when the bowser is inspected and deemed to be free of material presence.

9. The system as claimed in claim 1, wherein the alert generation unit (116) comprises communication channels for generating alerts through electronic communication, wherein the alerts are transmitted via the communication channels based on signals received from the AI server through the PLC.

10. A method (400) of bowser inspection using an automated inspection system (100), the method comprising:
mounting (402) a jib crane on a platform for bowser inspection, the jib crane having two degrees of freedom sliding motion both in vertical and horizontal direction;
serving (404) a motorized hoist (104) mounted on the jib crane (102) as a trolley for x-axis movement, the motorized hoist being equipped with a guider roller for smooth sliding movement;
operating (406), by a programmable logic controller (PLC) (110) integrated with an artificial intelligence (AI) server, upon receiving a set of data from a first set of sensors, a robotic arm aligns with the calculated position of the bowser hole, the set of data pertaining to comprehensive visual coverage and position of the browser hole, wherein the robotic arm (106) coupled to the motorized hoist and equipped with the first set of sensors and second set of sensors and adapted to move inside an inlet hole of the bowser to capture the set of images of the internal surface of the bowser;
analysing (408), by the PLC integrated with the AI server, the captured set of images received from the second set of sensors to perform image annotation by identifying class labels for the detected objects within the captured images, the set of images pertains to the internal surface of the bowser, wherein the analysed set of images is sorted to detect the presence and absence of the material within the bowser; and
emitting (410) an alert in response to signals received from the PLC indicating the presence of the material within the bowser, wherein an alert generation unit coupled to the PLC generates the alert for automated material detection in the bowser inspection.