Full-automatic automobile liquefied natural gas filling method and system based on mechanical arm

Abstract

The full-automatic automobile liquefied natural gas filling method based on the mechanical arm comprises the steps that S1, an automobile enters a gas filling station, and a gas filling instruction is sent to a liquid filling machine through a web server side; s2, the position of a gas filling port is collected, a collection result is processed, an interpolation pose is obtained, and the mechanical arm is in butt joint with the filling port according to the interpolation pose; s3, the liquid adding machine fills the automobile with liquefied natural gas, a mechanical arm is stabilized through rigidity impedance control and an adding control loop in the filling process, coordinates of the mechanical arm are monitored, and filling is stopped when abnormity occurs; and S4, after filling is completed, settlement and self-inspection are carried out, and if abnormity occurs, abnormity grades are divided, and response actions are executed. Accurate and effective automatic filling is achieved, manual operation errors are avoided, the probability of safety accidents is reduced, the filling process is simplified through automatic filling and block chain settlement, the filling time is shortened, and the operation efficiency of the gas station is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of LNG filling stations, and in particular to a fully automatic automobile liquefied natural gas filling method and system based on a robotic arm. Background Art

[0002] Currently, the filling of LNG (Liquefied Natural Gas) filling stations mainly relies on manual operation, which has the following disadvantages:

[0003] 1. Safety hazards: There is a risk of error in manual operation, which may lead to safety accidents such as LNG leakage, fire or even explosion, causing casualties and property losses.

[0004] 2. Inefficiency: Manual refueling is slow and can easily cause vehicle queues and congestion, affecting the operational efficiency of gas stations.

[0005] 3. Data silos: The refueling system and settlement system are independent of each other, and data cannot be communicated. This leads to inaccurate operational data statistics, making it difficult to conduct effective operational analysis and decision-making. Problems such as liquid outages are prone to occur, affecting customer service quality.

[0006] Furthermore, in order to reduce the burden of manual refueling, there is a method in the prior art for refueling automobile energy using a robotic arm. However, the existing robotic arm refueling method still has the following defects:

[0007] Traditional robotic arm control relies on a single sensor (such as vision or force perception), which can easily lead to positioning drift or force control failure in complex environments (such as reflections from the fuel filler port or slight vehicle movement).

[0008] The LNG filling port requires constant contact force, but slight deviations in the vehicle's parking position (±50 mm) and vibrations in the robotic arm can cause force control overshoot.

[0009] The timestamps of vision, force perception, and lidar data are not synchronized, resulting in image overlap and ghosting in fusion positioning.

[0010] The blockchain consensus delay (usually >1 second) does not match the real-time nature of the robotic arm's movements (requires a response time of <100ms).

[0011] LNG filling stations often face low temperature (-30°C) and high humidity (95% RH) environments, which can cause electronic components to fail. Summary of the Invention

[0012] The purpose of the present invention is to provide a fully automatic vehicle liquefied natural gas filling method based on a robotic arm, so as to solve the problem of low safety of existing natural gas filling.

[0013] In one aspect, an embodiment of the present invention provides a fully automatic vehicle liquefied natural gas refueling method based on a robotic arm, comprising:

[0014] S1, the car enters the gas station and sends a gas filling instruction to the liquid filling machine through the web server;

[0015] S2, collecting the position of the gas filling port and processing the collected results to obtain an interpolated posture, and docking the robot arm with the filling port according to the interpolated posture;

[0016] In step S3, the dispenser fills the vehicle with liquefied natural gas. During the filling process, the robot arm is stabilized through stiffness impedance control and an additional control loop. The robot arm coordinates are monitored and the filling is stopped when an abnormality occurs.

[0017] S4, after the filling is completed, settlement and self-check are carried out. If any abnormality occurs, the abnormality level is classified and the response action is executed.

[0018] Furthermore, the robot arm in S1 receives the gas filling instruction, uses the laser radar to correct the positioning and performs the gas filling port docking, and further includes:

[0019] S1.1: The vehicle enters the gas station and stops in the designated area next to the gas island;

[0020] S1.2: The user inputs the refueling information to the web server;

[0021] S1.3: The web server sends the refueling information to the LNG dispenser and transmits the refueling instruction to the robotic arm based on the refueling information.

[0022] Furthermore, the step of collecting the position of the gas filling port and processing the collected result to obtain an interpolated posture, and docking the robot arm with the filling port according to the interpolated posture in S2 further includes:

[0023] S2.1: After receiving the refueling command, the robotic arm synchronizes the clocks of the camera and lidar sensors, and collects the coordinates of the refueling port through the camera and lidar respectively;

[0024] S2.2, fuses the coordinates collected by the camera and lidar, and dynamically adjusts the confidence through the weighted covariance matrix;

[0025] S2.3, based on the frequency difference of the data collected by the camera and the lidar, the fusion results are calculated to obtain the interpolated pose.

[0026] Furthermore, synchronizing the clocks of the camera and the lidar sensors, and collecting the coordinates of the filling port through the camera and the lidar respectively, further includes:

[0027] Use PTP to synchronize the clocks of each sensor, so that the clock error of each sensor is less than 1ms. At the same time, the priority queue is set by default. When data conflicts, the lidar is used as the reference to correct visual positioning;

[0028] The coordinates of the filling port are located by SLAM using a camera and a lidar respectively, and the point cloud coordinates and radar coordinates of the filling port are obtained.

[0029] Furthermore, the step of calculating the fusion result to obtain the interpolated pose further includes:

[0030] By the time t of the visual frame cam Interpolation is used to obtain the corresponding lidar pose P lidar (t cam ) to achieve synchronization, assuming the timestamps of two adjacent frames of lidar data are t k and t k+1 , the corresponding poses are P k and P k+1 , then the visual time t cam ∈[t k ,t k+1 The interpolated pose of ] is:

[0031]

[0032] Where, P lidar (t cam ) represents the visual time t cam The interpolated pose of .

[0033] Furthermore, the S3 further includes:

[0034] S3.1: After the robotic arm is docked with the filling port, the command is sent back to the LNG dispenser, which starts to add liquid to the vehicle.

[0035] S3.2, dynamically changing stiffness impedance control during the filling process, adjusting impedance parameters in real time according to the end contact force;

[0036] S3.3, a band-stop filter is embedded in the control loop of the manipulator to reduce the vibration amplitude of the manipulator during the filling process;

[0037] S3.4, monitor the robot arm coordinates in real time during the filling process, and stop filling immediately if abnormal coordinates appear.

[0038] Furthermore, in S3.2, a mass-damping-stiffness model is constructed to adjust the impedance parameters, and the calculation formula is as follows:

[0039] F=M\ddot{x}+B\dot{x}+K(x-x0)

[0040] Among them, the stiffness coefficient K changes dynamically with the contact force error;

[0041] F represents the expected interaction force, that is, the target force that the robot wants to exert or track when it contacts the environment;

[0042] M represents the virtual mass matrix, which is used to characterize the inertial characteristics of the manipulator in the virtual impedance model;

[0043] ddot{x} represents the current acceleration;

[0044] B represents the virtual damping coefficient, which is used to characterize the damping characteristics of the manipulator in the virtual impedance model;

[0045] dotx indicates the current speed;

[0046] K represents the virtual stiffness coefficient, which is used to characterize the stiffness characteristics of the manipulator in the virtual impedance model;

[0047] x represents the current position;

[0048] x0 represents the desired position.

[0049] Furthermore, the S3.3 also includes:

[0050] By adding a band-stop filter to the controller and execution structure, the frequency domain of the vibration signal of the original robot arm satisfies:

[0051] x(t)=Asin(2πf v t)

[0052] Where x(t) represents the end vibration displacement, t = time variable, f v Represents the original vibration signal frequency domain, f v =80Hz, A represents the vibration amplitude, A=0.5mm;

[0053] The feedback signal of the vibration frequency is filtered out by adding a band-stop filter. The transfer function H(z) of the band-stop filter is as follows:

[0054]

[0055] Where H(z) represents the transfer function of the filter, which describes the response of the filter to signals of different frequencies.

[0056] z represents a complex variable, corresponding to the unit delay operator in discrete-time signals, z -1 Indicates the delay of one sampling period;

[0057] θ represents the normalized center frequency, which is determined by the vibration frequency f v and sampling frequency f s Calculate the normalized center frequency:

[0058]

[0059] r represents the pole radius (the value range is 0 < r < 1), which is used to control the bandwidth and attenuation depth of the band-stop filter.

[0060] When r approaches 1, it means that the stopband becomes narrower and the attenuation depth increases.

[0061] When r approaches 0, it means that the stopband becomes wider and the attenuation depth decreases.

[0062] The feedback signal of the vibration frequency is filtered out. When the coordinates at the end of the robotic arm change, it is transmitted to the controller through the feedback signal, and the pose is adjusted.

[0063] Furthermore, step S4 further includes:

[0064] S4.1, after the filling is completed, control the robotic arm to return to its original position and perform the fund settlement for this filling.

[0065] S4.2: Perform self-check after the gas filling is completed and execute periodic self-check. If an abnormality occurs, classify the abnormality level and execute the corresponding action.

[0066] On the other hand, an embodiment of the present invention further provides a full-automatic liquefied natural gas filling system for automobiles based on a robotic arm, including: a web server, an intelligent filling control module, a blockchain settlement module, and a self-check module;

[0067] The web server is used to send a gas filling instruction to the liquid filling machine after the automobile enters the gas filling station.

[0068] The intelligent filling control module is used to collect the position of the gas filling port, process the collected result to obtain an interpolated pose, and dock the robotic arm with the filling port according to the interpolated pose; control the liquid filling machine to fill the automobile with liquefied natural gas, stabilize the robotic arm through stiffness impedance control and adding a control loop during the filling process, and monitor the coordinates of the robotic arm. When an abnormality occurs, stop the filling.

[0069] The intelligent filling control module is used to perform settlement after the filling is completed.

[0070] The self-check module is used for self-check after the filling is completed and periodic self-check. When an abnormality occurs, classify the abnormality level and execute the corresponding action.

[0071] The beneficial effects of this invention include providing an intelligent refueling system and blockchain settlement platform for LNG refueling stations, aiming to improve the safety, efficiency, and operational management of refueling stations. This invention's innovation in the control algorithm for the robotic arm goes beyond simple motion programming. While designed with the specific needs of LNG refueling scenarios in mind, it also offers specific optimizations and breakthroughs in multimodal perception fusion and dynamic, compliant control. Secondly, in terms of the integration of multiple technologies, the present invention can achieve stable positioning in complex environments, with a significant improvement in accuracy compared to traditional single-sensor solutions; through the optimization of the dynamic impedance control strategy, the success rate of filling gun docking is increased to 99.7%, and the life of the sealing ring is extended; by adopting PTP (Precision Time Protocol) to synchronize the clocks of each sensor, the blockchain consensus delay error is less than 1ms; a priority queue is designed, and the laser radar's stronger anti-interference ability is utilized to correct the visual positioning based on the laser radar when data conflicts; a layered settlement architecture is designed to quickly complete pre-withholding through the local industrial computer, and the blockchain is asynchronously uploaded to the chain for evidence storage; through smart contract optimization, Fabric channel technology is used to separate settlement transactions from other data, the priority is raised to the "emergency channel", and the delay is compressed to 300ms; a self-test-fault-tolerant process is designed to switch to the redundant module when the sensor is abnormal (for example, after the laser radar fails, the pure visual positioning accuracy is degraded to ±3mm and it can still work);

[0072] Compared with the prior art, the present invention has the following advantages:

[0073] 1. High safety: Automated filling effectively avoids human operational errors and reduces the probability of safety accidents.

[0074] 2. Efficiency improvement: Automated filling and blockchain settlement simplify the filling process, shorten filling time, and improve gas station operating efficiency.

[0075] 3. Data transparency: Blockchain technology ensures the authenticity and immutability of data. Users and operators can query transaction records at any time, improving data transparency.

[0076] 4. Decision support: The data analysis module provides operators with accurate data support to help them optimize their operation strategies and improve operational efficiency and service quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0078] Figure 1 A schematic flow chart of a fully automatic vehicle liquefied natural gas refueling method based on a robotic arm provided in an embodiment of the present invention;

[0079] Figure 2 Schematic diagram of a conventional control circuit of a robotic arm in an embodiment of the present invention;

[0080] Figure 3 This is a schematic structural diagram of a fully automatic automobile liquefied natural gas filling system based on a robotic arm provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0081] To make the purpose, technical solutions and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0082] In the description of this specification, the terms "include", "including", "have", "contain", etc. are all open terms, which mean including but not limited to. The descriptions with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", etc. mean that the specific features, structures or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. The order of steps involved in each embodiment is used to schematically illustrate the implementation of the present application, and the order of steps therein is not limited and can be appropriately adjusted as needed.

[0083] Example 1

[0084] An embodiment of the present invention provides a fully automatic vehicle liquefied natural gas filling method based on a robotic arm, comprising the following steps:

[0085] Step 1: The car enters the gas station and sends a refueling instruction to the refueling machine through the web server;

[0086] The details are as follows:

[0087] Step 1.1: The vehicle enters the gas station and stops at the designated area next to the gas island. The area is guided by the solid line on the ground and the retractable stop sign installed above the canopy.

[0088] Step 1.2: The driver scans the QR code to start the gas filling app and selects the gas filling role as retail or fleet member.

[0089] Among them, if you choose to refuel at retail, you will need to obtain credit authorization;

[0090] Select Quota, Quantity, or Fill Up, confirm, and jump to step 1.3.

[0091] If you choose to refuel as a fleet member, enter the corresponding fleet account and password, retrieve and verify your identity from the web server, and return relevant fleet member information, balance information, etc.

[0092] Select Quota, Quantity, or Top Up. After confirmation, skip to step 1.3.

[0093] Step 1.3: The web server sends the refueling information to the LNG dispenser, which in turn transmits the refueling instruction to the robotic arm.

[0094] Step 2: Collect the position of the gas filling port and process the collected results to obtain the interpolated pose, and then dock the robot arm with the filling port according to the interpolated pose;

[0095] Step 2 further includes:

[0096] Step 2.1: After receiving the refueling command, the robotic arm synchronizes the clocks of the camera and lidar sensors, and collects the coordinates of the refueling port through the camera and lidar respectively;

[0097] Among them, synchronizing the clocks of the camera and lidar sensors includes:

[0098] PTP is used to synchronize the clocks of each sensor, so that the clock error of each sensor is less than 1ms. At the same time, the priority queue is set by default. When data conflicts, the lidar is used as the reference to correct visual positioning.

[0099] Furthermore, the coordinates of the filling port are located by SLAM using a camera and a lidar respectively, and the point cloud coordinates and radar coordinates of the filling port are obtained.

[0100] Preferably, the camera is an RGB-D camera.

[0101] Step 2.2: Fuse the coordinates collected by the camera and lidar, and dynamically adjust the confidence level through the weighted covariance matrix;

[0102] For example, when there is sufficient light, the visual weight is 70% and the lidar weight is 30%; the opposite is true in rainy and foggy days.

[0103] Step 2.3: Based on the frequency difference of the data collected by the camera and the lidar, the fusion results are calculated to obtain the interpolated pose;

[0104] The camera is a visual sensor. Due to the frequency difference between the visual (30Hz) and laser radar (20Hz) data, there is an error in the fusion positioning result. The present invention uses the time t of the visual frame to cam Interpolation is used to obtain the corresponding lidar pose P lidar (tcam ) to achieve synchronization:

[0105] By the time t of the visual frame cam Interpolation is used to obtain the corresponding lidar pose P lidar (t cam ) to achieve synchronization

[0106] Assume that the timestamps of two adjacent frames of lidar data are t k and t k+1 , the corresponding poses are P k and P k+1 , then the visual time t cam ∈[t k ,t k+1 The interpolated pose of ] is:

[0107]

[0108] Where, P lidar (t cam ) represents the visual time t cam The interpolated pose of .

[0109] Using the linear interpolation formula, both the error and the computational overhead tend to the intermediate value; after experimental verification, the positioning error after fusion can be reduced from ±3mm to ±1.2mm.

[0110] Based on the obtained interpolated pose, the robotic arm is docked with the filling port.

[0111] Step 3: The dispenser starts to add liquefied natural gas. During the filling process, the robot arm is stabilized through stiffness impedance control and an additional control loop, and the robot arm coordinates are monitored. If an abnormality occurs, the filling is stopped.

[0112] Step 3.1: After the robotic arm is docked with the filling port, the command is sent back to the LNG filling machine, and the LNG filling machine starts filling.

[0113] Step 3.2, dynamically changing stiffness impedance control during the filling process, adjusting impedance parameters in real time according to the end contact force;

[0114] Specifically, a mass-damping-stiffness model is constructed to adjust the impedance parameters. The calculation formula is as follows:

[0115] F=M\ddot{x}+B\dot{x}+K(x-x0)

[0116] Among them, the stiffness coefficient K changes dynamically with the contact force error;

[0117] F represents the expected interaction force, that is, the target force that the robot wants to exert or track when it contacts the environment;

[0118] M represents the virtual mass matrix, which is used to characterize the inertial characteristics of the manipulator in the virtual impedance model;

[0119] ddot{x} represents the current acceleration;

[0120] B represents the virtual damping coefficient, which is used to characterize the damping characteristics of the manipulator in the virtual impedance model;

[0121] dotx indicates the current speed;

[0122] K represents the virtual stiffness coefficient, which is used to characterize the stiffness characteristics of the manipulator in the virtual impedance model;

[0123] x represents the current position;

[0124] x0 represents the desired position;

[0125] Considering that the robotic arm may experience slight vibrations during the liquefied natural gas filling process, in order to avoid various problems such as changes in the coordinates of the robotic arm's completed operations and drift of the robotic arm's axes, the robotic arm's posture is slightly adjusted based on the calculated expected interaction force to ensure that the robotic arm is always in a flexible state.

[0126] Step 3.3: embed a band-stop filter in the control loop of the robotic arm to reduce the vibration amplitude of the robotic arm during the filling process;

[0127] By embedding a band-stop filter into the control loop of the robotic arm, the vibration amplitude of the robotic arm during the filling process was reduced from ±0.5mm to ±0.1mm.

[0128] Conventional control circuits of robotic arms are as follows Figure 2 As shown in the figure, when a fixed-frequency vibration (e.g., 80 Hz) occurs at the end of the robotic arm, the vibration energy is concentrated in a specific frequency band. By embedding a band-stop filter in the control loop, the control signal component corresponding to the vibration frequency is filtered out, preventing the controller from responding to the vibration frequency. This disrupts the transmission path of the vibration energy and suppresses mechanical resonance.

[0129] Preferably, a band-stop filter is added to the controller and the execution structure, and the original vibration signal frequency domain satisfies:

[0130] x(t)=Asin(2πf v t)

[0131] Where x(t) represents the end vibration displacement, t = time variable, f v Represents the original vibration signal frequency domain, f v =80Hz, A represents the vibration amplitude, A=0.5mm.

[0132] The transfer function H(z) of the band-stop filter is as follows:

[0133]

[0134] Among them, H(z) represents the transfer function of the filter (represented in the frequency domain or z-domain), which describes the response of the filter to signals of different frequencies.

[0135] z represents a complex variable, corresponding to the unit delay operator in the discrete-time signal (z -1 represents the delay of one sampling period);

[0136] θ represents the normalized center frequency, which is calculated from the vibration frequency f v and the sampling frequency f s as follows:

[0137]

[0138] r represents the pole radius (the value range is 0 < r < 1), which is used to control the bandwidth and attenuation depth of the band-stop filter.

[0139] r → 1 indicates that the stopband becomes narrower and the attenuation depth increases (the filtering is sharper).

[0140] r → 0 indicates that the stopband becomes wider and the attenuation depth decreases (the filtering is smoother)

[0141] Without adding a filter: The vibration signal is fed back to the controller through the sensor. The controller misjudges the existence of a position deviation and outputs a reverse control quantity, which may instead amplify the vibration (positive feedback).

[0142] After adding a band-stop filter: The feedback signal of the vibration frequency is filtered out. The controller no longer responds to the vibration (80 Hz) component of the fixed frequency at the end of the robotic arm, and the vibration energy cannot continuously accumulate in the closed loop.

[0143] When the coordinates at the end of the robotic arm change, it is transmitted to the controller through the feedback signal, and the pose is adjusted.

[0144] [ Step 3.4, monitor the coordinates of the robotic arm. If abnormal coordinates appear, stop the filling immediately;

[0145] Specifically, set the normal range value of the coordinates at the end of the robotic arm. If it exceeds the normal range value, it is marked as abnormal coordinates. Abnormal coordinates may be caused by external dragging, etc. If the sensor transmits abnormal coordinate information, a stop signal for the liquid filling machine is transmitted, and the controller performs a homing process.

[0146] Step 4, perform settlement and self-check after the filling is completed. If an abnormality occurs, classify the abnormality level and execute the corresponding action.

[0147] Step 4 also includes:

[0148] Step 4.1: After the filling is completed, the robot arm is controlled to return to its original position and the funds for this filling are settled;

[0149] The liquid filling machine conveys the filling completion and return instruction to the robotic arm.

[0150] The industrial computer generates withholding vouchers through the memory database.

[0151] The withholding data is stored in the local message queue and asynchronously uploaded to the chain.

[0152] The asynchronous on-chain process uses Fabric's emergency channel, using only necessary nodes. The endorsement policy for the settlement instruction only requires signatures from 2 / 3 of the institutions. After the on-chain confirmation, a confirmation signal is returned to the industrial computer, and the local ledger is released. (This part of the execution process is an asynchronous operation. After the robot arm returns to its position, the order is deemed to be completed.)

[0153] Fund security is guaranteed by the credit authorization collected during the initial QR code scan and the available balance obtained by fleet members. This avoids the drawbacks of conventional gas filling and cashier processes, such as the lack of a middleman and reliance on post-collection.

[0154] The settled data will be cleaned, integrated, and analyzed in a data pool to generate visual reports, making it easier for operators to conduct financial verification, analyze operational status, and optimize operational strategies.

[0155] Step 4.2: After the refueling is completed, self-check and periodic self-check are performed. If any abnormality occurs, the abnormality level is classified and the response action is performed.

[0156] Step 4.2.1: After refueling, the robot arm performs a quick self-check to check whether the hardware monitoring status meets the predetermined health value and prepare for the next refueling work in advance.

[0157] Step 4.2.2: Periodic self-test: During the shift handover process, an initialization restart self-test will be performed, including verification of sensors, actuators, communication lines, etc.

[0158] Step 4.2.3: When an abnormal error occurs, the fault level is divided according to the preset value and a response action is taken.

[0159] An example is shown in Table 1 below:

[0160] Table 1 Fault level and response action table

[0161] Fault level Fault Examples Response Action Mild Single sensor noise exceeds limit Logging, continuous observation Moderate Motor temperature is higher than 80°C Reduce running speed serious Joint self-test position deviation Emergency Stop

[0162] Based on the working mode of LNG filling stations, which operate 365x24, redundancy and degradation strategies are implemented in sensors, actuators, and communication systems.

[0163] The sensor uses a primary and backup encoder, with a voting mechanism that automatically switches to the backup encoder if the primary encoder fails. (The third person in the majority vote here is a virtual reference, based on the motor current and sensorless.)

[0164] The actuator is driven by dual motors. When one motor fails, the other motor takes over and limits the load.

[0165] The communication system uses dual CAN buses, and the channel switches to backup after three channel timeouts, while resending cached commands.

[0166] Example 2

[0167] like Figure 3 As shown, the present invention also proposes a fully automatic vehicle liquefied natural gas filling system based on a robotic arm, which is used to implement the fully automatic vehicle liquefied natural gas filling method based on a robotic arm described in Example 1. The system includes:

[0168] Intelligent filling control module:

[0169] High-precision robotic arm control algorithm is used to achieve precise positioning and motion control of the robotic arm.

[0170] Integrated 3D visual recognition technology to automatically identify vehicle information, fuel port location and fueling status.

[0171] Monitor the filling process in real time and automatically handle abnormal situations to ensure filling safety.

[0172] Blockchain settlement module:

[0173] Build a decentralized data storage and trading platform based on blockchain technology.

[0174] Smart contract technology is used to realize automatic chain uploading and settlement of injection data.

[0175] Provides a data query interface to facilitate users and operators to query transaction records and historical data.

[0176] Self-test module:

[0177] Clean, integrate and analyze the filling data and settlement data to achieve self-inspection of the filling process and periodicity.

[0178] Generate visual reports to provide data support for operators.

[0179] Leverage machine learning algorithms to predict gas station operating trends and optimize operational strategies.

[0180] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0181] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0182] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0183] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0184] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A fully automatic vehicle liquefied natural gas filling method based on a robotic arm, characterized in that: include: S1, the car enters the gas station and sends a gas filling instruction to the liquid filling machine through the web server; S2, collecting the position of the gas filling port and processing the collected results to obtain an interpolated posture, and docking the robot arm with the filling port according to the interpolated posture; In step S3, the dispenser fills the vehicle with liquefied natural gas. During the filling process, the robot arm is stabilized through stiffness impedance control and an additional control loop. The robot arm coordinates are monitored and the filling is stopped when an abnormality occurs. S4, after the filling is completed, settlement and self-check are carried out. If any abnormality occurs, the abnormality level is classified and the response action is executed.

2. The fully automatic vehicle liquefied natural gas filling method based on a robotic arm according to claim 1, characterized in that: include: In step S1, the robot arm receives the gas filling instruction, uses the laser radar to correct the positioning, and performs the gas filling port docking, which also includes: S1.1: The vehicle enters the gas station and stops in the designated area next to the gas island; S1.2: The user inputs the refueling information to the web server; S1.3: The web server sends the refueling information to the LNG dispenser and transmits the refueling instruction to the robotic arm based on the refueling information.

3. The fully automatic vehicle liquefied natural gas filling method based on a robotic arm according to claim 1 or 2, characterized in that: The step S2 of collecting the position of the gas filling port and processing the collected result to obtain an interpolated posture, and docking the robot arm with the filling port according to the interpolated posture also includes: S2.1: After receiving the refueling command, the robotic arm synchronizes the clocks of the camera and lidar sensors, and collects the coordinates of the refueling port through the camera and lidar respectively; S2.2, fuses the coordinates collected by the camera and lidar, and dynamically adjusts the confidence through the weighted covariance matrix; S2.3, based on the frequency difference of the data collected by the camera and the lidar, the fusion results are calculated to obtain the interpolated pose.

4. The fully automatic vehicle liquefied natural gas filling method based on a robotic arm as claimed in claim 3, characterized in that: The clock synchronization of the camera and the laser radar sensors, and the acquisition of the coordinates of the filling port by the camera and the laser radar respectively, further includes: Use PTP to synchronize the clocks of each sensor, so that the clock error of each sensor is less than 1ms. At the same time, the priority queue is set by default. When data conflicts, the lidar is used as the reference to correct visual positioning; The coordinates of the filling port are located by SLAM using a camera and a lidar respectively, and the point cloud coordinates and radar coordinates of the filling port are obtained.

5. The fully automatic vehicle liquefied natural gas filling method based on a robotic arm as claimed in claim 3, characterized in that: The step of calculating the interpolated pose from the fusion result further includes: By the time t of the visual frame cam The corresponding LiDAR pose is obtained by interpolation to achieve synchronization. Let the timestamps of the two adjacent LiDAR frames be t k and t k+1 , the corresponding poses are P k and P k+1 , then the visual time t cam ∈[t k ,t k+1 The interpolated pose of ] is: Where, P lidar (t cam ) represents the visual time t cam The interpolated pose of .

6. The fully automatic vehicle liquefied natural gas filling method based on a robotic arm according to claim 1, characterized in that: Said S3 further comprises: S3.1: After the robotic arm is docked with the filling port, the command is sent back to the LNG dispenser, which starts to add liquid to the vehicle. S3.2, dynamically changing stiffness impedance control during the filling process, adjusting impedance parameters in real time according to the end contact force; S3.3, a band-stop filter is embedded in the control loop of the manipulator to reduce the vibration amplitude of the manipulator during the filling process; S3.4, monitor the robot arm coordinates in real time during the filling process, and stop filling immediately if abnormal coordinates appear.

7. The fully automatic vehicle liquefied natural gas filling method based on a robotic arm according to claim 6, characterized in that: In S3.2, a mass-damping-stiffness model is constructed to adjust the impedance parameters, and the calculation formula is as follows: F=M\ddot{x}+B\dot{x}+K(x-x0) Among them, the stiffness coefficient K changes dynamically with the contact force error; F represents the expected interaction force, that is, the target force that the robot wants to exert or track when it contacts the environment; M represents the virtual mass matrix, which is used to characterize the inertial characteristics of the manipulator in the virtual impedance model; \(\ddot{x}\) represents the current acceleration; B represents the virtual damping coefficient, which is used to characterize the damping characteristics of the robotic arm in the virtual impedance model; \(\dot{x}\) represents the current velocity; K represents the virtual stiffness coefficient, which is used to characterize the stiffness characteristics of the robotic arm in the virtual impedance model; x represents the current position; x0 represents the desired position.

8. The fully automatic vehicle liquefied natural gas filling method based on a robotic arm according to claim 6, characterized in that: The S3.3 further includes: Adding a band-stop filter to the controller and the execution structure, and the vibration signal frequency domain of the original robotic arm satisfies: x(t)=Asin(2πf v t) Where x(t) represents the end vibration displacement, t = time variable, f v Represents the original vibration signal frequency domain, f v =80Hz, A represents the vibration amplitude, A=0.5mm; Filtering the feedback signal of the vibration frequency by adding a band-stop filter, and the transfer function H(z) of the band-stop filter is as follows: where H(z) represents the transfer function of the filter, which describes the response of the filter to signals of different frequencies; z represents a complex variable, corresponding to the unit delay operator in discrete-time signals, z -1 Represents the delay of one sampling period; θ represents the normalized center frequency, which is determined by the vibration frequency f v and sampling frequency f s Calculate the normalized center frequency: r represents the pole radius (the value range is 0 < r < 1), which is used to control the bandwidth and attenuation depth of the band-stop filter. r approaching 1 means that the stopband becomes narrower and the attenuation depth increases, r approaching 0 means that the stopband becomes wider and the attenuation depth decreases; The feedback signal of the vibration frequency is filtered. When the coordinates at the end of the robotic arm change, it is transmitted to the controller through the feedback signal, and the pose is adjusted.

9. The fully automatic vehicle liquefied natural gas filling method based on a robotic arm according to claim 1, characterized in that: The S4 further includes: S4.1, after the filling is completed, control the robotic arm to return to its original position and perform the capital settlement for this filling; S4.2: Perform self-check after the gas filling is completed and execute periodic self-check. If an abnormality occurs, classify the abnormality level and execute the corresponding action.

10. A fully automatic automobile liquefied natural gas filling system based on a robotic arm, characterized in that: It includes: a web server, an intelligent filling control module, a blockchain settlement module, and a self-check module; The web server is used to send a gas filling instruction to the liquid filling machine after the vehicle enters the gas filling station; The intelligent filling control module is used to collect the position of the gas filling port, process the collected result to obtain the interpolation pose, and dock the robotic arm with the filling port according to the interpolation pose; control the liquid filling machine to fill the vehicle with liquefied natural gas, stabilize the robotic arm through stiffness impedance control and adding a control loop during the filling process, and monitor the coordinates of the robotic arm. When an abnormality occurs, stop the filling; The intelligent filling control module is used to perform settlement after the filling is completed; The self-check module is used for self-check after the filling is completed and periodic self-check, classify the abnormality level and execute the corresponding action when an abnormality occurs.

Images