Remote control system based on courier station robot arm

Abstract

The invention belongs to the technical field of post automation control, and particularly relates to a post robot arm-based remote control system, which comprises a local control unit, an encrypted data interaction gateway, a remote control center, a robot arm and a post transport vehicle, the local control unit integrates multiple sensors to acquire operation data and preprocesses the operation data, and the data is transmitted to the remote control center through an encryption link; the remote control center constructs and maintains an operation characteristic data set, analyzes data deviation through a multi-machine cooperative monitoring module, generates optimization parameters and performs graded early warning; the robot arm is of a six-degree-of-freedom series structure, the joint movement range can be remotely calibrated, and tail end execution parameters and tracks are dynamically optimized. The system can dynamically adapt to goods categories and operation height, guarantee safe and continuous operation, improve sorting efficiency and reduce energy consumption.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of automatic control of relay stations, and particularly relates to a relay station robot arm remote control system. BACKGROUND

[0002] With the rapid development of the e-commerce logistics industry, the cargo processing capacity of express relay stations is increasing, and the problems of low efficiency and high cost of traditional manual operation are becoming increasingly prominent. In the prior art, some relay stations have adopted robot arms to realize automatic sorting of goods, but there are still the following defects: on the one hand, the current robot arm control system is not adaptive to scene factors such as real-time cargo quantity changes, cargo weight and shape differences of relay stations, and cannot dynamically adjust the control strategy according to the scene differences, resulting in large fluctuations in sorting efficiency and difficulty in meeting the intelligent operation needs of complex logistics scenes; on the other hand, the processing mechanism for abnormal operation of the robot arm has obvious shortcomings, and over-intervention for slight abnormalities may cause operation interruption, and delayed response to serious abnormalities may cause equipment damage or goods damage risk, and the overall operation stability is insufficient. SUMMARY

[0003] In view of the above deficiencies in the prior art, the present application provides a relay station robot arm remote control system to solve the problems in the background art.

[0004] In order to solve the above technical problems, the present application adopts the following technical solutions: The relay station robot arm remote control system comprises a local control unit, a data interaction gateway, a remote control center, a robot arm and a relay station transport vehicle carrying the robot arm. The local control unit is in communication connection with the robot arm, is used for controlling the robot arm and integrating a plurality of types of sensor modules to collect original multi-dimensional data in the operation process of the robot arm, the content of which includes physical characteristics of goods, motion energy consumption data at different height positions, real-time end execution parameters, joint motion trajectories and abnormal processing signals, and the original multi-dimensional data is cleaned and preliminarily classified; The data interaction gateway adopts an encrypted communication link, is used for real-time transmission of the multi-dimensional data preprocessed by the local control unit to the remote control center, and simultaneously receives the regulation and control instructions issued by the remote control center; The remote control center constructs and maintains a machine arm operation characteristic data set based on the received multi-dimensional data, the machine arm operation characteristic data set adopts an incremental dynamic updating mechanism, and the content includes physical characteristics of goods generated through statistical analysis, motion energy consumption baselines at different height positions, historical end execution parameter benchmark values, joint motion trajectory models, and standardized abnormality processing logs; the remote control center is also provided with a multi-machine cooperative monitoring module for real-time analysis of the deviation of the operation data of each machine arm from the machine arm operation characteristic data set, generation of optimization parameters and pushing to the local control unit, and hierarchical early warning and intervention on abnormal states; The machine arm adopts a 6-degree-of-freedom series structure, and the motion range of each joint can be independently calibrated by the remote control center; and the end execution parameters and motion trajectories of the machine arm are dynamically iteratively optimized based on the optimization parameters pushed by the remote control center.

[0005] Preferably, the multi-type sensor module comprises: The depth camera mounted on the top of the relay transport vehicle is used to collect the physical characteristics of the goods; The current sensor integrated in the machine arm driving motor loop is used to collect motion energy consumption data at different operation height positions; The six-dimensional force sensor and the weight sensor built-in in the end effector of the machine arm are used to collect the clamping force, force feedback signal and weight data of the goods, respectively, to form the end execution parameters; The incremental encoder installed at each joint of the machine arm is used to collect real-time angle and rotation speed data of the joint to generate a joint motion trajectory; The MEMS acceleration sensor installed on the machine arm joint and the relay transport vehicle chassis is used to collect joint jitter instantaneous acceleration and transport vehicle movement vibration data.

[0006] Further, the remote control center generates a characteristic baseline based on the machine arm operation characteristic data set; the triggering condition of the incremental dynamic updating mechanism is that the Euclidean distance deviation of new operation data from the characteristic baseline exceeds a threshold value, or a new type of goods packaging appears, or a new operation height position is added; and the process of the incremental dynamic updating mechanism is as follows: S1, data cleaning: the local control unit removes abnormal operation data in the motion energy consumption data and the end execution parameters collected by the multi-type sensor module to obtain valid operation data; S2, feature matching: the remote control center compares the physical characteristics of the goods and the operation height position in the valid operation data with the corresponding dimensional data in the characteristic baseline through a cosine similarity algorithm to identify whether they are operation data of the same type of goods at the same height position; S3, Label Completion: If the cargo is identified as a new type of cargo or a new height location, the cargo's vulnerability level and height location value are marked through the human-machine interface of the remote control center. After the labeling is completed, the label is associated with the valid operational data. S4, Baseline Update: The remote control center uses a weighted fusion algorithm to correct the feature baseline, updates the average clamping force, energy consumption baseline and mean value of joint motion parameters at the corresponding height position of similar goods, and stores the valid operation data of the associated tags in the robot arm operation feature dataset. S5, Data Synchronization: The updated feature baseline is automatically pushed to the local control unit of all robotic arms in the same station.

[0007] 11. Furthermore, the multi-machine collaborative monitoring module's hierarchical early warning mechanism targets abnormal robotic arm operations, auxiliary equipment malfunctions, and communication anomalies, specifically including: Level 1 warning: The standard deviation a of the joint vibration acceleration of the robotic arm is greater than 0.6 m / s². 2 The remote control center pushes a warning message to the local control unit and automatically fine-tunes the joint damping parameters and movement speed, while triggering the local control unit's grasping retry mechanism. The remote control center also pushes a warning message to the local control unit and automatically fine-tunes the joint damping parameters and movement speed. The remote control center also pushes a warning message to the local control unit and automatically fine-tunes the joint damping parameters and movement speed. The remote control center also pushes a warning message to the local control unit and triggers the grasping retry mechanism. Level 2 warning: The standard deviation a of the joint vibration acceleration of the robotic arm is greater than 0.8 m / s². 2 The following conditions must be met: the duration of the failure is >5s and ≤10s; or the energy consumption is 20% to 30% higher than the baseline of motion energy consumption for the same type of goods and the same working height; or the grab fails 3 times in a row; or the data transmission interruption lasts >3s and ≤10s; or the vibration acceleration of the station transport vehicle exceeds the warning threshold and lasts >5s; the remote control center triggers a force control parameter forced correction command. Level 3 warning: The standard deviation a of the joint vibration acceleration of the robotic arm is greater than 1.0 m / s². 2 If the following conditions are met: the duration of the failure is ≥10s, or the energy consumption is more than 30% higher than the baseline energy consumption of the same type of goods and the same working height position; or there are 5 consecutive failed grabbing attempts; or the data transmission interruption lasts for more than 10s; or the vibration acceleration of the station transport vehicle exceeds the danger threshold and lasts for more than 10s; the remote control center immediately sends an emergency stop command, locks the robotic arm and the station transport vehicle, and pushes an audible and visual alarm to the remote monitoring terminal, which will be unlocked after manual investigation.

[0008] Furthermore, the standard deviation of joint vibration acceleration is calculated based on the instantaneous acceleration collected by the MEMS accelerometer, the average acceleration within the period, and the number of samplings.

[0009] Further, the dynamic iterative optimization method of the end execution parameter is: taking the average clamping force of the same goods category and the same operation height position in the robot arm operation characteristic data set as the reference, combining the shape deviation and weight deviation of the current goods and the characteristic baseline, introducing a stability weight coefficient to calculate the optimized clamping force; the shape deviation is calculated based on the three-dimensional profile of the goods collected by the depth camera, and is divided into regular cubes, irregular packages and soft package goods according to the goods packaging form; the weight deviation is the relative deviation of the current goods weight and the average weight of the characteristic baseline.

[0010] Further, the dynamic iterative optimization method of the motion trajectory is: taking the average energy consumption baseline of the same goods category and the same operation height position in the robot arm operation characteristic data set as the reference, combining the height deviation of the current operation height and the characteristic baseline, introducing an energy-saving weight coefficient to calculate the optimized motion energy consumption; the height deviation is the relative deviation of the current operation height and the average height of the characteristic baseline.

[0011] Further, the stability weight coefficient and the energy-saving weight coefficient are dynamically adjusted based on the stability score; the stability score is calculated by weighting multi-dimensional operation data; the multi-dimensional operation data specifically includes the clamping success rate, the clamping force stability and the joint motion stability of the robot arm.

[0012] Further, the remote control center calls the robot arm operation characteristic data set to generate a three-dimensional mapping table containing the goods category, the operation height position and the optimal operation parameter; when the cosine similarity of the to-be-operated goods and the same sample in the robot arm operation characteristic data set is ≥ 90% and the operation height is within the characteristic baseline coverage range, the remote control center automatically pushes the historical optimal operation parameter to the local control unit, including the clamping force threshold range, the optimal interval of the grabbing angle and the low-energy motion trajectory parameter.

[0013] Preferably, the relay transport vehicle realizes autonomous obstacle avoidance through laser radar, and the moving speed is linked with the robot arm operation, that is, when the operation height of the robot arm rises, the moving speed of the relay transport vehicle decreases.

[0014] Compared with the prior art, the present application has the following beneficial effects: 1. By constructing the robot arm operation characteristic data set and adopting the incremental dynamic updating mechanism, the system can accurately match the operation demand of different goods categories and operation height positions. At the same time, based on the dynamic iterative optimization of the data set, the historical optimal parameter can be automatically pushed, so that the adaptation efficiency of the robot arm to new goods is greatly improved, and the scene of the relay station with various goods categories and variable operation height is effectively coped with.

[0015] 2. The hierarchical early warning mechanism of the multi-machine cooperative monitoring module covers robot arm operation abnormalities and communication abnormalities, and each level of early warning matches differentiated intervention measures: first-level early warning automatically adjusts joint parameters and triggers a retry, second-level early warning limits operation height and corrects force control parameters, third-level early warning urgently locks the device and pushes a sound and light alarm, forming a complete protection chain of "early warning-intervention-recovery", avoiding the interruption of the entire operation caused by a single abnormality, and reducing the risk of goods damage and equipment failure.

[0016] 3. The dynamic iterative optimization method of the motion trajectory adjusts the operation height and the baseline deviation of energy consumption, saving the consumption of electricity; the end execution parameter optimization dynamically adjusts the clamping force according to the actual situation of the goods, avoiding unnecessary energy consumption. BRIEF DESCRIPTION OF DRAWINGS

[0017] Fig. 1 It is a structural schematic diagram of the relay station robot arm remote control system based on the application. Fig. 2 It is a flowchart of the incremental dynamic updating mechanism of the application. DETAILED DESCRIPTION

[0018] In order to enable those skilled in the art to better understand the application, the technical solutions of the application will be further described below in combination with the drawings and examples.

[0019] Among them, the drawings are only used for illustrative explanation, and the representation is only a schematic diagram, not a physical diagram, and cannot be understood as a limitation on the application; in order to better illustrate the embodiments of the application, some components in the drawings will be omitted, enlarged or reduced, and do not represent the size of the actual product; for those skilled in the art, it is understandable that some known structures and their descriptions in the drawings may be omitted.

[0020] The same or similar reference numerals in the drawings of the embodiments of the application correspond to the same or similar components; in the description of the application, it should be understood that if the terms "upper", "lower", "left", "right", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore the terms describing the positional relationship in the drawings are only used for illustrative explanation, and cannot be understood as a limitation on the application, for those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0021] In the description of this invention, unless otherwise explicitly specified and limited, the term "connection" or similar designation indicating a connection between components should be interpreted broadly. For example, it can refer to a fixed connection, a detachable connection, or an integral part; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can refer to the internal communication between two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. Example 1:

[0022] like Figs. 1-2 As shown, this invention addresses the shortcomings of existing robotic arm remote control systems, which suffer from insufficient adaptability to various scenarios and untimely anomaly handling. The core components of this system include a local control unit, a data interaction gateway, a remote control center, a robotic arm, and a transport vehicle carrying the robotic arm. These modules work together to achieve a closed-loop process of "data acquisition - transmission - analysis and optimization - execution monitoring," thereby enhancing the intelligence and stability of robotic arm operations.

[0023] The local control unit adopts an industrial-grade embedded controller, which communicates with the 6-DOF serial joint drive system of the robotic arm via a CAN bus. On the one hand, it analyzes the pulse signals fed back by the joint encoder in real time to control the arm's movements. On the other hand, it integrates multiple types of sensor modules to complete the acquisition and preliminary processing of multi-dimensional data during the operation.

[0024] The data interaction gateway has a built-in AES-256 data encryption protocol and supports 5G / WiFi6 dual-mode communication: it prioritizes WiFi6 connection to the station's local area network, and automatically switches to 5G when the network is unstable to ensure the security and stability of data transmission between the local and remote ends.

[0025] The remote control center is deployed on a cloud server and adopts a distributed architecture, which combines big data storage, algorithm analysis and multi-machine collaborative scheduling capabilities.

[0026] The robotic arm uses a 6-DOF serial structure, and the range of motion of each joint can be independently set and calibrated through dedicated calibration software in a remote control center.

[0027] The top of the transport vehicle is equipped with a fixed mounting base for the robotic arm, and the bottom integrates a lidar and drive wheel set, providing the robotic arm with a platform for mobile operations and obstacle avoidance capabilities.

[0028] The multi-type sensor modules are the core carriers for the system to perceive the operation scene, and the configuration and function of each sensor are closely adapted to the operation needs of the station.

[0029] The incremental encoder installed at each joint of the robot arm acquires the pulse signals of the joint rotation in real time, converts the real-time angle and rotation speed of the joint, and further generates continuous joint motion trajectory data, providing original data for subsequent motion trajectory optimization and ensuring that the trajectory adjustment has accurate historical motion characteristics to refer to. The six-axis force sensor and weight sensor built into the end effector of the robot arm can acquire force and torque signals in X, Y, and Z directions, accurately identify the size of the clamping force and abnormal fluctuations in force feedback, and directly obtain the weight of the goods. The combination of the two forms a set of end execution parameters of "clamping force-weight-force feedback", providing data support for subsequent clamping force optimization. The depth camera mounted on the top of the relay transport vehicle captures three-dimensional images of the goods and extracts the contours to obtain the length, width, height, and surface material characteristics of the goods, and then combines the weight sensor data to completely construct the physical characteristics of the goods. The MEMS acceleration sensor installed on the joints of the robot arm and the chassis of the relay transport vehicle has a sampling frequency of 100 Hz, which acquires instantaneous acceleration of joint jitter and transport vehicle movement vibration data, providing a signal source for jitter abnormality judgment. The current sensor integrated in the motor loop of the robot arm detects the motor operating current to calculate the energy consumption, combines with the work height position information, and generates a motion energy consumption baseline at different height positions, laying the foundation for energy consumption optimization.

[0030] Based on the pre-processed multi-dimensional data uploaded by the local control unit, the remote control center constructs and maintains a machine arm operation characteristic data set, which covers "goods physical characteristics (three-dimensional contour, weight, material, etc.), motion energy consumption baseline at different height positions, historical end execution parameters, joint motion trajectory, standardized abnormality processing log", etc., and continuously optimizes using an incremental dynamic update mechanism.

[0031] The feature baseline generation needs to meet the condition that the effective sample size of a single product category and a single operation height position is accumulated ≥ 50 groups, and the generation process is as follows: First, the local control unit cleans the operation data collected by the multi-type sensor module. Specifically, according to the type of goods and the operation height position, the data is classified, and the average energy consumption and force feedback signal fluctuation range of each category (same type of goods, same operation height position) are calculated. For energy consumption, if the energy consumption collected by the current sensor of the motor loop of the robot arm in a certain group of operation data exceeds 25% of the average energy consumption of the corresponding category, the group of data is excluded. For force feedback signals, when the force feedback signal collected by the six-axis force sensor built into the end effector of the robot arm fluctuates more than ± 30%, the group of data is also excluded. The pre-processed data is uploaded to the remote control center through the data interaction gateway.

[0032] Then, the remote control center receives the uploaded data and further classifies them according to the dimensions of goods category and working height. For each category, relevant statistical parameters are calculated. For example, the average clamping force F 基线 is obtained by calculating the arithmetic mean of multiple effective clamping force data collected by the six-axis force sensor and weight sensor built in the end effector; the average motion energy consumption E 基线 is calculated according to the energy consumption data collected by the drive motor loop current sensor, combined with the working height position information (provided by the joint encoder or depth camera); the average joint motion angle and standard deviation are obtained by statistical analysis of the real-time angle data collected by the incremental encoder installed at each joint of the robot arm. These statistical parameters form the initial feature parameter set together.

[0033] Then, the remote control center performs cross-device verification through the multi-machine cooperative monitoring module. The multi-machine cooperative monitoring module compares the data features of different robot arms under the same goods category and working height position. For example, it analyzes the consistency of joint motion angle, clamping force, energy consumption, etc. of multiple robot arms grabbing the same type of goods at the same height. If the differences in the same data features of different robot arms are within a reasonable range (for example, the difference in the average clamping force of each robot arm is not more than 10%, the difference in the average motion energy consumption is not more than 15%, etc., the specific threshold can be set according to the actual working scene and device performance), the verification is passed, and these initial feature parameter sets are determined as the official feature baseline.

[0034] Finally, the verified feature baseline is pushed to all local control units for storage, serving as the reference for subsequent data comparison and optimization during the robot arm working process. It is worth noting that when the system is started for the first time (without historical data), the initial baseline is generated based on trial operation data. During the trial operation stage, the robot arm is arranged to perform multiple grabbing operations on different goods categories and different working height positions. The local control unit collects the operation data and uploads it to the remote control center, which calculates and generates the initial baseline according to the above similar process (but the sample size requirement can be appropriately reduced, such as setting it to 30 groups) to provide an initial reference standard for subsequent formal operation.

[0035] The triggering conditions of the incremental dynamic updating mechanism include three scenarios: The first is that the Euclidean distance deviation between new operation data and feature baseline exceeds 15%, specifically: select the weight of goods, three-dimensional contour size (length, width, height), energy consumption as the feature dimension, first standardize each dimension data, that is, eliminate unit difference, then calculate the overall difference value of new operation data and feature baseline through the Euclidean distance formula, when the value exceeds 15% of the normal fluctuation range of the baseline, it is determined that the data distribution difference is significant; the second is the appearance of new type of goods packaging, such as originally processing express paper box, adding foam box or cloth bag packaging; the third is the addition of operation height position, such as the addition of 2.5m high shelf layer in the relay station. Its update process is divided into five steps: S1, data cleaning: the local control unit eliminates abnormal operation data in the motion energy consumption data and end execution parameters collected by the multi-type sensor module, obtains effective operation data and uploads to the remote control center; S2, feature matching: the remote control center compares the physical characteristics of goods and operation height position in the effective operation data with the corresponding dimension data in the feature baseline through the cosine similarity algorithm, to identify whether it is the operation data of the same type of goods at the same height position; S3, label completion: if it is determined to be a new type of goods or a new height position, mark the goods damage level (such as fragile, ordinary, pressure-resistant) and height position value through the human-computer interaction interface of the remote control center, and after the label is completed, the label is associated with the effective operation data; S4, baseline update: the remote control center uses weighted fusion algorithm to correct the feature baseline, updates the average clamping force, energy consumption baseline and joint motion parameter mean of the corresponding height position of the same type of goods, and stores the effective operation data associated with the label in the robot arm operation feature data set; S5, data synchronization: the updated feature baseline is automatically pushed to the local control unit of all robot arms in the same relay station, realizing the coordination and consistency of multi-machine feature baseline, supporting the collaborative optimization of multiple devices.

[0036] The multi-machine collaborative monitoring module of the remote control center adopts a hierarchical early warning mechanism, covering three scenarios of robot arm operation abnormality, auxiliary equipment abnormality and communication abnormality, and each warning level has clear trigger standard and intervention measures. Specifically: The triggering scenarios of the first level warning include: the joint jitter acceleration standard deviation a of the robot arm is greater than 0.6m / s 2 , and the duration is less than or equal to 5s, the calculation formula of the joint jitter acceleration standard deviation is , where a iThe instantaneous acceleration collected by the MEMS acceleration sensor is a, the average acceleration in the period is a, and n is the number of samples; the energy consumption is higher than the baseline of the same working condition motion energy consumption by 10% to 20%, the energy consumption data collected by the current sensor is compared with the baseline; single grabbing failure, whether the goods are still attached to the end of the robot arm is identified through the depth camera, and if not, it is determined as failure; the duration of data transmission interruption is ≤3s, which is judged by the communication state detection module of the data interaction gateway. After triggering the first level warning, the remote control center pushes the text and light warning information to the local control unit, automatically adjusts the joint damping parameters and movement speed, and triggers the retry mechanism of the local control unit to re-plan the grabbing trajectory and execute the grabbing action again.

[0037] The triggering scenario of the second level warning includes: the joint jitter acceleration standard deviation a of the robot arm is >0.8m / s 2 , and the duration is >5s and ≤10s; the energy consumption is higher than the baseline of the same working condition motion energy consumption by 20% to 30%; the number of grabbing failures within 1 minute is counted by the remote control center; the duration of data transmission interruption is >3s and ≤10s; the moving vibration acceleration of the relay transport vehicle exceeds the warning threshold and the duration is >5s. After triggering the second level warning, the remote control center sends a force control parameter forced correction instruction, which reduces the fluctuation range of the end gripping force from ±15% to ±10%, limits the operation height to below 1.5m, and temporarily suspends the movement of the relay transport vehicle to avoid the influence of transport vehicle vibration on the stability of robot arm operation.

[0038] The triggering scenario of the third level warning includes: the joint jitter acceleration standard deviation a of the robot arm is >1.0m / s 2 , and the duration is ≥10s; the energy consumption is higher than the baseline of the same working condition motion energy consumption by more than 30%; the number of grabbing failures is 5 times in a row; the duration of data transmission interruption is >10s; the moving vibration acceleration of the relay transport vehicle exceeds the dangerous threshold and the duration is ≥10s. After triggering the third level warning, the remote control center immediately sends an emergency stop instruction to cut off the power supply of the robot arm drive motor and lock the drive wheel of the relay transport vehicle through the CAN bus; at the same time, an audible and visual alarm is pushed to the remote monitoring terminal, and after the management personnel arrive at the scene to troubleshoot the fault (such as checking sensor wiring and repairing communication link), an unlock password is input at the remote control center, and the system can resume normal operation.

[0039] The end execution parameters of the robot arm are dynamically iteratively optimized based on the robot arm operation characteristic data set to balance "stability" and "energy consumption". The optimization method takes the average gripping force F 基线 of the same goods category and the same operation height position in the robot arm operation characteristic data set as the reference, calculates the optimized gripping force F combining the shape deviation ΔS and weight deviation ΔW of the current goods and the characteristic baseline, and the formula is: , Where k1 and k2 are stability weight coefficients, whose values ​​are dynamically adjusted by the stability score; ∆S is calculated based on the 3D contour of the goods captured by the depth camera, and is divided into three categories according to the type of goods packaging: ∆S is 0 for regular cubes; ∆S is 0.1~0.3 for irregular packages; and ∆S is 0.3~0.5 for flexible packaging; the formula for calculating ∆W is: For example, if the average weight of similar goods in the characteristic baseline is 5kg and the current weight of the goods is 6kg, then ∆W = (6−5) / 5 = 0.2.

[0040] This optimization method can adjust the clamping force according to the actual shape and weight of the goods, avoiding the goods from falling off due to excessively loose clamping or the goods packaging being damaged due to excessively tight clamping. It is especially suitable for scenarios where there are a variety of goods in the station.

[0041] Dynamic iterative optimization of motion trajectory is based on the average energy consumption E of the same product category and the same working height position in the robot arm operation feature dataset. 基线 Based on the baseline, the optimal motion energy consumption E is calculated by combining the height deviation ∆H between the current working height and the characteristic baseline. The formula is as follows: , Where k3 is the energy-saving weighting coefficient, which is also dynamically adjusted by the stability score; the formula for calculating ∆H is: , For example, if the average operating height of similar goods in the characteristic baseline is 1.8m and the current operating height is 1.5m, then ∆H = (1.8 − 1.5) / 1.8 ≈ 0.167.

[0042] This optimization method can adjust the motion energy consumption according to the working height, appropriately reduce energy consumption when working at height (such as reducing the output power of the motor), and at the same time ensure the smoothness of the trajectory, so as to avoid increasing operating costs due to excessive energy consumption or affecting the stability of cargo grasping due to trajectory fluctuations.

[0043] The stability weighting coefficients k1 and k2, and the energy-saving weighting coefficient k3 are adjusted based on the stability score, which is calculated by weighting multi-dimensional operational data. The formula is as follows:

[0044] Where S is the stability score, ranging from 0 to 10; R is the grab success rate, ranging from 0 to 1, which is the ratio of the number of successful grabs to the total number of grabs per unit time. For example, if 8 out of 10 grabs are successful, then R = 0.8. This refers to the fluctuation range of the end-grip force, which is the difference between the maximum and minimum values ​​of the gripping force during the grasping process, such as F. 基线 =100N, the clamping force fluctuates between 95N and 105N, then δ F =10N; The joint angle fluctuation amplitude, i.e., the maximum deviation value of the actual joint movement angle from the reference angle, is denoted as θ 基线 = 2° (the reference joint angle fluctuation amplitude of the same type of operation trajectory), the joint angle fluctuation is 1°-3°, and δ θ = 2°; ω1, ω2, and ω3 are index weights, and The weight distribution is based on the scenario characteristics of the relay station operation, i.e., “loss due to failure to grab > energy consumption cost”, and by default, ω1=0.5 (highlighting the core position of the successful clamping rate), ω2=0.3 (emphasizing the stability of the clamping force), and ω3=0.2 (considering the stability of the joint movement). When the stability score S<7, it indicates that the current operation stability is low, and the grabbing safety needs to be prioritized, at this time, k1=0.3, k2=0.2, and k3=0.1 are set; when the stability score S≥9, it indicates that the operation stability is good, and the energy saving optimization can be focused on, at this time, k1=0.1, k2=0.1, and k3=0.3 are set, and the dynamic balance between stability and energy consumption is achieved through weight adjustment.

[0045] The remote control center also has a three-dimensional mapping table generation and parameter pushing function, which calls the robot arm operation characteristic data set to generate a three-dimensional mapping table containing the goods category, operation height position, and optimal operation parameters. The goods category is divided into express cartons, foam boxes, cloth bags, plastic boxes, etc., the height position is divided into five layers according to the relay station shelf layer, i.e., 0.7 m (sorting table), 1.2 m, 1.8 m, 2.0 m, and 2.5 m, and the optimal parameters are selected from the parameters with a stability score ≥9 in the historical operation (such as clamping force threshold, grabbing angle range, and low-energy consumption movement trajectory parameters) and stored in the MySQL database. When a new goods enters the operation area, the depth camera collects its three-dimensional characteristics and weight, and matches the samples in the robot arm operation characteristic data set for cosine similarity: when calculating the cosine similarity, the weight of the goods weight is 0.4, and the weight of the length / width / height is 0.2. If the matching degree is ≥90% and the operation height is within the characteristic baseline coverage range, the remote control center automatically pushes the corresponding historical optimal parameters to the local control unit. For example, when matching to the “express carton-1.8 m” category, the parameters of the clamping force threshold 80-120 N, the grabbing angle 30°-45°, and the joint movement angular velocity limit 10° / s are pushed, without the need for manual debugging, the operation can be quickly started, and the adaptation efficiency of new goods is greatly improved.

[0046] The relay station transport vehicle realizes autonomous obstacle avoidance through a laser radar, the scanning range of the laser radar is 360°, the detection distance is 0.1-10 m, and the surrounding obstacles (such as shelves, personnel, and other equipment) are scanned in real time. When the distance of the detected obstacle is ≤1 m, the transport vehicle automatically reduces the moving speed from 1.2 m / s to 0.6 m / s; when the distance is ≤0.5 m, the transport vehicle stops moving, and resumes operation after the obstacle is removed.

[0047] Meanwhile, the moving speed of the transport vehicle is linked with the working height of the robot arm, and the current working height of the robot arm is obtained through the communication between the local control unit and the transport vehicle controller to dynamically adjust the vehicle speed: when the working height is less than or equal to 1.5 m, the moving speed of the transport vehicle is set to 1 m / s; when the working height is greater than 1.5 m and less than or equal to 2.0 m, the speed is reduced to 0.8 m / s; and when the working height is greater than 2.0 m, the speed is further reduced to 0.5 m / s. By reducing the speed of the transport vehicle during high-altitude operation, the influence of vehicle body vibration on the end of the robot arm is reduced, and the stability of high-altitude grabbing is ensured.

[0048] To sum up, the embodiment realizes remote control and intelligent optimization of the relay station robot arm through multi-sensor fusion perception, data set incremental update, hierarchical early warning intervention, parameter dynamic optimization and multi-device cooperative linkage, and solves the problems of insufficient scene adaptability and incomplete abnormal handling in the prior art.

[0049] The above is only an embodiment of the present application, and the circuit and electronic components and modules involved are prior art, which can be realized by those skilled in the art without further description. The content protected by the present application does not involve improvement of software and methods. The specific structure and characteristics of the scheme known in the art are not described in detail here. Those skilled in the art know all the ordinary technical knowledge in the field of the application as of the filing date or the priority date, can obtain all the prior art in the field, and have the ability to apply conventional experimental means before the date. Those skilled in the art can improve and implement the present scheme based on their own ability under the guidance of the present application. Some typical known structures or known methods should not be an obstacle for those skilled in the art to implement the present application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present application, which should be considered as the protection scope of the present application. These will not affect the effect and practicality of the present application.

Claims

1. A remote control system for a robotic arm at a rest stop, characterized in that: Includes a local control unit, a data interaction gateway, a remote control center, a robotic arm, and a transport vehicle that carries the robotic arm; The local control unit is communicatively connected to the robotic arm and is used to control the robotic arm and integrate multiple types of sensor modules to collect raw multi-dimensional data during the operation of the robotic arm. The data includes the physical characteristics of the cargo, motion energy consumption data at different height positions, real-time end effector parameters, joint motion trajectories and abnormal handling signals. At the same time, the raw multi-dimensional data is cleaned and preliminarily classified. The data interaction gateway uses an encrypted communication link to transmit multi-dimensional data preprocessed by the local control unit to the remote control center in real time, while receiving control commands issued by the remote control center. The remote control center constructs and maintains a robotic arm operation feature dataset based on the received multi-dimensional data. The robotic arm operation feature dataset adopts an incremental dynamic update mechanism and includes the physical characteristics of the cargo generated by statistical analysis, the motion energy consumption baseline at different height positions, the historical end-effector execution parameter benchmark values, the joint motion trajectory model, and the standardized anomaly handling log. The remote control center is also equipped with a multi-machine collaborative monitoring module, which is used to analyze the deviation between the operation data of each robotic arm and the robotic arm operation feature dataset in real time, generate optimization parameters and push them to the local control unit, and at the same time provide graded early warning and intervention for abnormal states. The robotic arm adopts a 6-DOF serial structure, and the range of motion of each joint can be independently calibrated by a remote control center; and the end effector parameters and motion trajectory of the robotic arm are dynamically iteratively optimized based on the optimization parameters pushed by the remote control center.

2. The remote control system for a robotic arm based on a rest stop as described in claim 1, characterized in that: The multi-type sensor module includes: The depth camera mounted on the top of the station transport vehicle is used to collect the physical characteristics of the goods; The current sensor integrated into the drive motor circuit of the robotic arm is used to collect motion energy consumption data at different working heights. The robotic arm's end effector has a built-in six-dimensional force sensor and weight sensor that collect the clamping force, force feedback signal, and cargo weight data to form end effector parameters. The incremental encoders installed at each joint of the robotic arm are used to collect real-time joint angle and rotation speed data and generate joint motion trajectories. The MEMS accelerometers installed on the robotic arm joints and the chassis of the transport vehicle are used to collect instantaneous acceleration data of joint vibration and vibration data of the transport vehicle movement.

3. The remote control system for a robotic arm based on a rest stop as described in claim 1, characterized in that: The remote control center generates a feature baseline based on the robotic arm's operational feature dataset; the incremental dynamic update mechanism is triggered when the Euclidean distance deviation between the new operational data and the feature baseline exceeds a threshold, or when a new type of cargo packaging appears, or when a new operational height position is added; the process of the incremental dynamic update mechanism is as follows: S1, Data Cleaning: The local control unit removes abnormal operation data from the motion energy consumption data collected by various types of sensor modules and the end execution parameters to obtain valid operation data; S2, Feature Matching: The remote control center uses a cosine similarity algorithm to compare the physical features of the goods and the height position of the operation in the effective operation data with the corresponding dimension data in the feature baseline to identify whether the operation data are for the same type of goods at the same height position. S3, Label Completion: If the cargo is identified as a new type of cargo or a new height location, the cargo's vulnerability level and height location value are marked through the human-machine interface of the remote control center. After the labeling is completed, the label is associated with the valid operational data. S4, Baseline Update: The remote control center uses a weighted fusion algorithm to correct the feature baseline, updates the average clamping force, energy consumption baseline and mean value of joint motion parameters at the corresponding height position of similar goods, and stores the valid operation data of the associated tags in the robot arm operation feature dataset. S5, Data Synchronization: The updated feature baseline is automatically pushed to the local control unit of all robotic arms in the same station.

4. The remote control system for a robotic arm based on a rest stop as described in claim 1, characterized in that: The hierarchical early warning mechanism of the multi-machine collaborative monitoring module targets abnormal robot arm operation, auxiliary equipment malfunctions, and communication anomalies, specifically including: Level 1 warning: The standard deviation a of the joint vibration acceleration of the robotic arm is greater than 0.6 m / s². 2 The remote control center pushes warning information to the local control unit and automatically fine-tunes the joint damping parameters and movement speed, while triggering the local control unit's grasping retry mechanism. The remote control center also pushes warning information to the local control unit and automatically fine-tunes the joint damping parameters and movement speed. The remote control center also pushes warning information to the local control unit and automatically fine-tunes the joint damping parameters and movement speed. The remote control center also pushes warning information to the local control unit and triggers the grasping retry mechanism. Level 2 warning: The standard deviation a of the joint vibration acceleration of the robotic arm is greater than 0.8 m / s². 2 The following conditions must be met: the duration of the failure is >5s and ≤10s; or the energy consumption is 20% to 30% higher than the baseline of motion energy consumption for the same type of goods and the same working height; or the grab fails 3 times in a row; or the data transmission interruption lasts >3s and ≤10s; or the vibration acceleration of the station transport vehicle exceeds the warning threshold and lasts >5s; the remote control center triggers a force control parameter forced correction command. Level 3 warning: The standard deviation a of the joint vibration acceleration of the robotic arm is greater than 1.0 m / s². 2 If the following conditions are met: the duration of the failure is ≥10s, or the energy consumption is more than 30% higher than the baseline energy consumption of the same type of goods and the same working height position; or there are 5 consecutive failed grabbing attempts; or the data transmission interruption lasts for more than 10s; or the vibration acceleration of the station transport vehicle exceeds the danger threshold and lasts for more than 10s; the remote control center immediately sends an emergency stop command, locks the robotic arm and the station transport vehicle, and pushes an audible and visual alarm to the remote monitoring terminal, which will be unlocked after manual investigation.

5. The remote control system for a robotic arm based on a rest stop as described in claim 4, characterized in that: The standard deviation of the joint vibration acceleration is calculated based on the instantaneous acceleration collected by the MEMS accelerometer, the average acceleration within the period, and the number of samplings.

6. The remote control system for a robotic arm based on a rest stop as described in claim 1, characterized in that: The dynamic iterative optimization method for the end-effector execution parameters is as follows: taking the average clamping force of the same type of goods and the same working height position in the robot arm operation feature dataset as the benchmark, and combining the shape deviation and weight deviation of the current goods from the feature baseline, a stability weight coefficient is introduced to calculate and optimize the clamping force; the shape deviation is calculated based on the three-dimensional contour of the goods collected by the depth camera, and is divided into regular cubes, irregular packages and soft packaging goods according to the packaging form; the weight deviation is the relative deviation between the current weight of the goods and the average weight of the feature baseline.

7. The remote control system for a robotic arm based on a rest stop as described in claim 1, characterized in that: The dynamic iterative optimization method for the motion trajectory is as follows: taking the average energy consumption baseline of the same cargo category and the same working height position in the robot arm operation feature dataset as the benchmark, and combining the height deviation between the current working height and the feature baseline, an energy-saving weight coefficient is introduced to calculate and optimize the motion energy consumption; the height deviation is the relative deviation between the current working height and the average height of the feature baseline.

8. The remote control system for a station-based robotic arm as described in claim 6 or 7, characterized in that: The stability weighting coefficient and energy-saving weighting coefficient are dynamically adjusted based on the stability score; the stability score is calculated by weighting multi-dimensional operation data; the multi-dimensional operation data specifically includes gripping success rate, gripping force stability, and joint motion stability of the robotic arm.

9. The remote control system for a robotic arm based on a rest stop as described in claim 1, characterized in that: The remote control center calls the robotic arm operation feature dataset to generate a three-dimensional mapping table containing cargo category, operation height position, and optimal operation parameters. When the cosine similarity between the cargo to be operated and the same type of sample in the robotic arm operation feature dataset is ≥90% and the operation height is within the coverage range of the feature baseline, the remote control center automatically pushes the historical optimal operation parameters to the local control unit, including the clamping force threshold range, the optimal gripping angle range, and low-energy motion trajectory parameters.

10. The remote control system for a robotic arm based on a rest stop as described in claim 1, characterized in that: The station transport vehicle achieves autonomous obstacle avoidance through lidar, and its movement speed is linked to the operation of the robotic arm. That is, when the working height of the robotic arm increases, the movement speed of the station transport vehicle decreases accordingly.

Images