Automatic airport luggage sorting system and method based on self-adaptive degree-of-freedom mechanical arm

Abstract

The invention discloses an automatic airport luggage sorting system and method based on a self-adaptive degree-of-freedom mechanical arm. The system comprises a mechanical arm body, a sensing module, a control module and an interface module. The mechanical arm body comprises six rigidity-adjustable joints, a pneumatic-electromagnetic hybrid driving design is adopted, and the degree of freedom can be dynamically adjusted; the sensing module collects luggage state and environment information in real time; the control module operates a reinforcement learning algorithm based on a Transform architecture; and the interface module exchanges data with an airport central control system through the ROS 2 middleware. The system solves the problems that a traditional mechanical arm is low in efficiency and insufficient in adaptability, luggage processing efficiency and accuracy are improved, and deployment cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of airport baggage sorting, and in particular to an automatic airport baggage sorting system and method based on an adaptive degree of freedom robotic arm. Background Art

[0002] With increasingly busy airport operations, the efficiency and precision of baggage handling systems have become critical factors. Traditional baggage handling systems primarily rely on fixed-degree-of-freedom robotic arms or manual operations. While these systems can accomplish tasks such as baggage sorting, transfer, and loading to a certain extent, they suffer from significant efficiency bottlenecks and insufficient adaptability. While fixed-degree-of-freedom robotic arms can perform basic grasping tasks, they struggle with the flexible movement requirements of complex environments, such as avoiding obstacles or performing high-precision baggage stacking operations, impacting overall handling efficiency. Furthermore, manual operations rely on high skills and long, repetitive workloads, which are not only costly but also prone to errors and injuries due to fatigue.

[0003] To address these issues, existing technologies attempt to enhance system flexibility and adaptability by improving control strategies. For example, rule-based algorithms are employed to address the diverse baggage load and the dynamic nature of the handling environment. However, these strategies often appear relatively simplistic when faced with complex and ever-changing real-world situations, lacking flexibility and the ability to respond to emergencies. Consequently, existing systems still lack integration between robotic arm operation, sensor monitoring, and baggage management systems, leading to delayed system responses or accumulated errors, requiring manual intervention to address sorting errors.

[0004] In summary, existing baggage handling systems suffer from numerous limitations, including fixed degrees of freedom, a single control strategy, and low system integration. A new technical solution is urgently needed that balances flexibility and precision, adapts to diverse environments, and aims to improve the overall efficiency and safety of airport baggage handling systems. In light of this, the present invention addresses the numerous shortcomings and inconveniences caused by the imperfections of existing baggage handling systems, and has been developed through extensive research, improvements, and trial production. Summary of the Invention

[0005] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide an airport baggage automatic sorting system and airport baggage automatic sorting method based on an adaptive degree of freedom robotic arm that can balance flexibility and precision and adapt to various environments, thereby improving the overall efficiency and safety of airport baggage handling.

[0006] In order to achieve the above object, the solution of the present invention is: An airport baggage automatic sorting system based on an adaptive degree of freedom robotic arm comprises: a robotic arm body, a sensor module, a control module and an interface module; The robotic arm body adopts a modular design, comprising multiple joints with adjustable stiffness and multiple connecting rods, each joint being connected to a control module; The sensing module includes an FBG sensor, a visual camera, and an RFID reader. The FBG sensor is embedded in the robotic arm body and connected to the control module. The visual camera is used to obtain luggage posture information, including the location and posture information of the luggage; the RFID reader is used to identify the luggage tag. The FBG sensor, visual camera, and RFID reader are respectively connected to the control module. The control module runs a reinforcement learning algorithm based on the Transformer architecture, dynamically adjusts the processing order based on baggage priority and system load, and outputs joint stiffness commands and corresponding drive commands to the robotic arm body; The interface module is integrated into the robotic arm body. The connector module exchanges data with the airport's baggage management system via the ROS 2 middleware connected to the control module. The data includes input information, including baggage trajectory, baggage priority, and system status, and output information, including loading status feedback.

[0007] Furthermore, the joint adopts a pneumatic-electromagnetic hybrid drive mode, and the stiffness of the joint Determined by the following formula: = × + × ; in, is the air pressure value of the jth joint, ranging from 0 to 1 MPa; is the electromagnetic coil current of the jth joint, ranging from 0 to 5A; is the aerodynamic stiffness coefficient, which depends on the diaphragm material and structure of the joint; is the electromagnetic stiffness coefficient, which is related to the performance of the permanent magnet and the number of coil turns; When the knuckles are locked, the critical value of the electromagnetic coil current; when ≥ , the joint enters a locked state and the robot's degrees of freedom are reduced.

[0008] Furthermore, the policy function π of the reinforcement learning algorithm is defined as =π( ), in, is the state observation at time t, including the luggage posture, conveyor belt speed v, and the joint angles and strain data of the robot arm; is the action output, including the and Set value; θ is the policy network parameter, which is obtained by minimizing the loss function ζ: θ = argmin ζ( , ), Argmin represents the set of independent variable values ​​that makes the function achieve the minimum value; Indicates that the discount factor is 0.99; The loss function ζ comprehensively considers task completion time, energy consumption, and operation accuracy: ζ = + ‖ ‖ The weight coefficient , , Determined through domain knowledge, Refers to the position of the luggage, Refers to the target posture of the luggage.

[0009] Furthermore, the input information of the interface module includes baggage trajectory =[x,y,z,α,β,η]; x,y,z represent the three-dimensional coordinates of the luggage on the conveyor belt, α is the yaw angle of the luggage, β is the pitch angle of the luggage, and η is the roll angle of the luggage; Priority flag f∈{0,1}; when f=0, it indicates low-priority baggage, and when f=1, it indicates high-priority baggage; System status code c∈{0,...,255}; a binary numerical identifier indicating the overall operating status of the airport baggage handling system, with a value range of 0 to 255; The output information is the loading status feedback s∈{0,1}, and when s=0, the system-level error correction process is triggered.

[0010] Furthermore, the robotic arm body includes 6 joints with adjustable stiffness, and the connecting rod is made of carbon fiber reinforced polymer.

[0011] Furthermore, the FBG sensor in the sensing module is embedded in the carbon fiber reinforced polymer connecting rod of the robotic arm body, tightly integrated with the connecting rod, and strains synchronously with the deformation of the connecting rod. The FBG sensor monitors the status of the robotic arm body in real time. By detecting the strain data of the connecting rod, it indirectly obtains the force conditions, motion posture and potential vibration risks of each joint of the robotic arm body, providing closed-loop feedback for the control algorithm. When the robotic arm grasps the body to pick up luggage of different weights or shapes, the FBG sensor provides real-time feedback on the connecting rod strain, assisting the control module in determining whether the joint stiffness needs to be adjusted, thereby achieving dynamic adjustment of the degrees of freedom.

[0012] Furthermore, the baggage management system and the ROS 2 middleware implement data interaction through a collaborative optimization protocol, which dynamically allocates sorting tasks based on task priority and system load.

[0013] Furthermore, the robotic arm body, sensor module, control module and interface module are connected via wired or wireless means to achieve real-time transmission of data and execution of control instructions.

[0014] The present invention can also adopt the following technical solutions: An automatic airport baggage sorting method according to the automatic airport baggage sorting system based on the adaptive degree of freedom robotic arm comprises the following steps: S1, Baggage Transfer Stage: When a bag is transferred into the sensor module's collection range, the RFID reader acquires the baggage information, and the visual camera captures the baggage's precise position and posture. The collected data is then sent to the control module, which calculates the optimal grasping path based on the received data and configures the robot arm's degrees of freedom by adjusting the joint stiffness. S2, during the baggage grabbing phase, the robotic arm maintains a high degree of freedom configuration to ensure flexibility and grabs the baggage according to the optimal grabbing path calculated by the control module; During the baggage handling and placement phase, the robotic arm grabs the bag and switches to a low-degree-of-freedom configuration to improve accuracy and stability, moving and placing the baggage along the optimal path. During the entire process, the FBG sensor continuously monitors the strain state of the robotic arm body and provides real-time feedback for the control algorithm.

[0015] Furthermore, when faced with luggage of different weights and shapes, the robotic arm body automatically adjusts the air pressure and current values ​​of the joints according to the guidance of the strategy function.

[0016] After adopting the above solution, the airport baggage automatic sorting system and method based on the adaptive degree of freedom robotic arm of the present invention has the following advantages compared with the existing technology: 1. The present invention's robotic arm utilizes a hybrid drive design for multiple joints, featuring a pioneering pneumatic-electromagnetic composite stiffness adjustment mechanism. Each joint utilizes a pneumatic-electromagnetic hybrid drive system, overcoming the response speed limitations of traditional single-drive systems and enabling rapid stiffness adjustment. The joints can adjust their stiffness characteristics in real time based on task requirements, addressing the limitations of traditional robotic arms' degrees of freedom and control strategies and improving the present invention's adaptability to complex baggage sorting scenarios. When grasping irregularly shaped luggage, joint stiffness can be reduced to enhance maneuverability; when handling and placing luggage, joint stiffness can be increased to enhance stability. This real-time adjustment based on task requirements creates a programmable stiffness curve. Joint stiffness adjustment can be performed in real time during the robotic arm's operation. While conventional robotic arms have fixed joint characteristics or are difficult to adjust, the present invention can adapt joint characteristics in real time based on luggage status (position, shape, priority) and environmental conditions (conveyor speed, space constraints), enabling proactive reconfiguration. For example, when encountering luggage of varying sizes and shapes, or when conveyor speeds fluctuate, the robotic arm can rapidly adjust joint stiffness and degrees of freedom, optimizing its motion strategy and improving baggage handling efficiency and accuracy.

[0017] 2. The robotic arm of the present invention adopts dynamic degree of freedom management, which can adjust the degree of freedom of the robotic arm in real time based on task requirements. The FBG sensor is used to monitor the vibration state of the robotic arm in real time and feedback it to the control module, so that the control strategy can be adjusted in time, improving the adaptability and stability of the robotic arm, and ensuring efficient and accurate baggage sorting.

[0018] 3. The control module of the present invention adopts an intelligent control architecture, integrates a hybrid decision-making model of Transformer and reinforcement learning, and designs a policy function considering the dynamic characteristics of the robot body. The joint stiffness and degrees of freedom of the robot body have a significant impact on its dynamic characteristics, and the policy function is closely related to these factors. When the joint stiffness and degrees of freedom are adjusted according to the task requirements, the policy function will change accordingly. During the grasping phase, the robot arm needs to maintain a high degree of freedom configuration to ensure flexibility. At this time, the policy function will tend to give more rewards to flexible grasping actions, such as allowing appropriate increases in energy consumption and time within a certain range to successfully grasp luggage with irregular shapes or uncertain positions. During the carrying and placement phase, the robot arm switches to a low degree of freedom configuration to improve accuracy and stability. The policy function focuses more on positioning accuracy and reducing vibration, giving higher rewards to precise placement actions, while penalizing behaviors that may lead to reduced accuracy, such as excessive joint movement amplitude or unstable posture.

[0019] A reinforcement learning algorithm based on the Transformer architecture optimizes the control strategy of the robot arm using a policy function. During training, the agent receives reward feedback based on its actions in different states and continuously adjusts its policy to maximize the cumulative reward. Because the policy function accounts for the dynamic characteristics of the robot arm, the agent gradually learns to select actions that conform to the laws of dynamics during learning. When faced with luggage of varying weights and shapes, the agent automatically adjusts the air pressure and current values ​​of its joints, guided by the policy function, to achieve optimal grasping and handling while meeting energy efficiency and accuracy requirements.

[0020] 4. System-level innovations of this invention include the integration of FBG sensors with the manipulator, which continuously monitors the arm's strain state and provides real-time feedback for the control algorithm. An interface module serves as a bridge for external communication, enabling coordinated optimization of the baggage management system and the manipulator. The integration of adjustable stiffness joints (pneumatic-electromagnetic hybrid drive) with carbon fiber connecting rods, and the integration of carbon fiber connecting rods with FBG sensors, provides the physical foundation for dynamic degree of freedom adjustment. This represents the hardware. A distributed communication framework based on ROS 2 enables real-time data exchange between the manipulator and the central system, ensuring seamless integration of sensor data, control instructions, and task scheduling. This represents the software architecture. A reinforcement learning model integrates the manipulator's dynamic characteristics with system-level constraints (baggage priority, conveyor speed, etc.) to generate a globally optimal strategy for algorithm optimization. This deep synergy of hardware, software, and algorithms significantly improves the efficiency and accuracy of baggage handling.

[0021] In summary, this invention proposes an innovative automated airport baggage sorting system and method based on an adaptive degree-of-freedom (DOF) robotic arm. Designed specifically for automated airport baggage sorting and loading scenarios, the system's core innovation lies in its dynamic DOF adjustment capability and intelligent control strategy. By adjusting the robotic arm's joint stiffness and DOF configuration in real time, it achieves seamless switching between flexible grasping and precise placement modes. This dynamic DOF adjustment allows the system to quickly adapt to baggage of varying sizes and shapes, resolving the problem of grasping failures often associated with traditional robotic arms due to their fixed structure. The intelligent control strategy dynamically adjusts the processing order based on baggage priority and system load, optimizing overall throughput. Furthermore, the system seamlessly integrates with existing baggage management infrastructure, significantly reducing deployment costs while ensuring system reliability and maintainability. Compared to traditional fixed-DOF robotic arms, the system's motion strategy automatically optimizes based on the baggage's real-time status (e.g., location, shape, priority) and environmental conditions (e.g., conveyor speed, space constraints), significantly improving baggage handling efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1This is a schematic diagram of the integration of the airport baggage handling system and the adaptive freedom robotic arm of the present invention.

[0023] Figure 2 Schematic diagram of the structure of the adaptive degree of freedom robotic arm system of the present invention. DETAILED DESCRIPTION

[0024] In order to further explain the technical solution of the present invention, the present invention is described in detail below through specific embodiments.

[0025] like Figure 1 and Figure 2 As shown, the present invention discloses an automatic airport baggage sorting system based on an adaptive degree of freedom robotic arm. The system is connected to a baggage handling system and a central management system. The baggage handling system is used to handle airport baggage and includes a check-in counter, a baggage drop-off point, a conveyor system, and security equipment. The check-in counter is used for the initial check-in process for passengers and generates baggage tag information. The baggage drop-off point is a self-service baggage check-in device that delivers baggage to the conveyor system. The conveyor system transports baggage to core conveyor equipment at different processing stages. The security equipment scans baggage to ensure compliance with aviation safety standards.

[0026] The central management system is a central control unit for airport baggage handling.

[0027] The invention discloses an automatic airport baggage sorting system based on an adaptive degree of freedom mechanical arm, comprising a mechanical arm body, a sensor module, a control module and an interface module.

[0028] The robot body adopts a modular design, including six adjustable stiffness joint mechanisms and carbon fiber reinforced polymer connecting rods. The carbon fiber reinforced polymer connecting rods are lightweight and high-strength robot arm structural components; the adjustable stiffness joint mechanism is the key executive component to achieve dynamic adjustment of the robot arm's degree of freedom. Each joint adopts a pneumatic-electromagnetic hybrid drive method, breaking through the response speed limit of the traditional single drive method and achieving rapid stiffness adjustment. The pneumatic-electromagnetic hybrid drive of the present invention is an innovative drive mechanism, and the stiffness of the joint is Determined by the following formula: = × + ×

[0029] in, is the air pressure value of the j-th joint (0-1MPa); is the electromagnetic coil current of the jth joint (0-5A); is the aerodynamic stiffness coefficient, which depends on the diaphragm material and structure; is the electromagnetic stiffness coefficient, which is related to the performance of the permanent magnet and the number of coil turns; When the knuckles are locked, the critical value of the electromagnetic coil current; when ≥ When the joint enters the locked state, the robot arm's degrees of freedom are reduced accordingly.

[0030] This design enables a single joint to switch from fully flexible to fully rigid within 10ms, providing the hardware foundation for real-time adjustment of the robot's degrees of freedom.

[0031] The sensing module is responsible for collecting baggage status and environmental information in real time, including FBG sensors, visual cameras and RFID readers.

[0032] Fiber Bragg Grating (FBG) sensors are embedded in the carbon fiber reinforced polymer (CFRP) connecting rods of the robotic arm. They are tightly integrated with the connecting rod material and strain in sync with the connecting rod's deformation. The FBG sensors monitor the robotic arm's status in real time. By detecting strain data on the connecting rod, they indirectly determine the force applied to each joint, its motion posture, and potential vibration risks, providing closed-loop feedback for the control algorithm. When the robotic arm grasps luggage of varying weights or shapes, the FBG sensors provide real-time feedback on the connecting rod strain, helping the control module determine whether joint stiffness needs to be adjusted (for example, switching from a flexible state with high degrees of freedom to a rigid state with low degrees of freedom). This ensures grasping stability and motion accuracy, and assists in dynamic adjustment of degrees of freedom. By monitoring abnormal fluctuations in strain data over a long period of time, fatigue or damage to the robotic arm's structure can be detected in advance, reducing the risk of system failure and improving maintainability.

[0033] Visual cameras and RFID readers can be deployed in the conveyor system or its periphery. The visual cameras are used to obtain luggage posture information, including the location and posture information of the luggage; the RFID readers are used to identify the luggage tags.

[0034] The control module is the intelligent decision-making center of the present invention. It runs a reinforcement learning algorithm based on the Transformer architecture, dynamically adjusts the processing order based on baggage priority and system load, optimizes overall throughput, and outputs joint stiffness commands and pneumatic valve commands to the robot body. The joint stiffness commands are control signals output to the robot body, while the pneumatic valve commands are precise control commands for adjusting air pressure.

[0035] The policy function π of the Transformer-based reinforcement learning algorithm (i.e., the agent) is defined as =π( ), in, is the state observation at time t, including the luggage posture, conveyor belt speed v, the angles and strain data of each joint of the robot arm, etc.; is the action output, including the and Setting value θ is the policy network parameter, which is obtained by minimizing the loss function ζ: θ = argmin ζ( , ), Argmin represents the set of independent variable values ​​that makes the function achieve the minimum value; represents the discount factor, which is set to 0.99; The loss function ζ comprehensively considers task completion time, energy consumption, and operation accuracy: ζ = + ‖ ‖ The weight coefficient , , Determined through domain knowledge, Refers to the position of the luggage, Refers to the target posture of the luggage.

[0036] Domain knowledge refers to professional experience, industry rules and actual needs in airport baggage sorting scenarios. The task completion time is based on the logic that airport baggage handling must meet the flight schedule. Overtime may result in baggage missing or flight delays. The energy consumption (air pressure + current) is directly affected by the energy consumption of the pneumatic system and electromagnetic coil, especially in large-scale deployment (such as deploying hundreds of robotic arms in a single airport). Balance performance and energy consumption; For operation accuracy, the value logic is that insufficient luggage placement accuracy may cause conveyor belt blockage or loading errors (such as luggage sliding out of the sorting grid). W3 is required to ensure that the error is less than the safety threshold.

[0037] , , The initial weights (e.g. = 0.5:0.3:0.2), giving priority to ensuring sorting efficiency and accuracy. , , It can also be optimized through on-site testing: During the trial operation phase at the airport, the coefficient can be adjusted based on measured data (such as sorting success rate and energy consumption curve under different weights). For example, if the risk of robotic arm overload is high in a certain scenario, the coefficient can be temporarily increased. To reduce joint stiffness (reduce current / air pressure output).

[0038] The combination of weight coefficients and domain knowledge reflects the algorithm's adaptation from "theoretical model" to "engineering implementation", ensuring that the control strategy meets the actual operational needs of the airport and avoiding infeasible solutions caused by purely data-driven methods, such as sacrificing the sorting time of all baggage in pursuit of accuracy.

[0039] Discount Factor The present value coefficient used to measure future rewards, with a value range of [0,1]. In this invention, the discount factor The value of 0.99 is mainly based on long-term profit orientation, stability considerations and industry practice reference. When it is close to 1, it means that the agent attaches more importance to long-term cumulative rewards, which is suitable for scenarios that require multi-step planning (such as optimizing the robot's handling path). In airport baggage sorting, the robot needs to take into account the coordination of the current grasping task and subsequent tasks (such as avoiding the inability to execute subsequent tasks due to excessive current energy consumption). Therefore, it is necessary to encourage the algorithm to pursue the long-term optimal strategy. A larger discount factor This reduces the strategy's sensitivity to immediate noise (such as small fluctuations in conveyor belt speed), making the robot's movements smoother and more predictable, meeting the high reliability requirements of airport systems. The proposed discount factor of 0.99 prioritizes long-term system stability.

[0040] The value of the discount factor enhances the system's long-term stability and multi-task coordination capabilities. It is a key parameter supporting the "intelligent control architecture" to achieve dynamic degree of freedom optimization, and forms a logical closed loop with the innovative point of "adjusting the strategy based on baggage priority and system load" in this invention.

[0041] Through the above design, the control algorithm of the present invention can not only quickly respond to real-time conditions (such as avoiding sudden obstacles), but also optimize overall performance from a global perspective, significantly improving the intelligence level of the airport baggage sorting system.

[0042] The interface module is a bridge for the system of the present invention to communicate with the outside world, and exchanges data with the baggage management system through the ROS 2 middleware. The ROS 2 middleware is a communication protocol for realizing integration with the airport management system.

[0043] The input information includes: Baggage track =[x,y,z,α,β,η]; x,y,z represent the three-dimensional coordinates of the luggage on the conveyor belt, α is the yaw angle of the luggage, β is the pitch angle of the luggage, and η is the roll angle of the luggage.

[0044] Priority flag f∈{0,1}; when f=0, it indicates low-priority baggage, and when f=1, it indicates high-priority baggage; System status code c∈{0,...,255}: A binary numerical identifier representing the overall operating status of the airport baggage handling system, with a value range of 0 to 255 (a total of 256 states).

[0045] The output information is loading status feedback s∈{0,1}. When s=0, the system-level error correction process is triggered. Through automated fault detection, multi-level response, and algorithm optimization, this loading status feedback and system-level error correction process establishes a complete closed loop from "anomaly perception" to "recovery execution," improving the efficiency and reliability of airport baggage handling.

[0046] Baggage trajectory parameters =[x,y,z,α,β,η] is the core input of the reinforcement learning algorithm. It determines the shape complexity of the luggage through the posture parameters and drives the joint stiffness formula. = × + × The optimal stiffness configuration is calculated; the central management system dynamically assigns sorting tasks to the nearest robotic arm based on position parameters x, y and priority (f) through ROS 2. Comprehensive perception of six baggage trajectory parameters enables accurate modeling of baggage status, providing a solid data foundation for the intelligent operation of the robotic arm.

[0047] The interface module enables collaborative optimization between the baggage management system and the present invention, ensuring efficient operation of the airport automatic baggage sorting system. The ROS 2 middleware enables collaboration between the two, including data exchange, task allocation, and system feedback and error correction.

[0048] Data Interaction: ROS 2 middleware bridges the data exchange between the baggage management system and the robotic arm, enabling the control module to receive information from the baggage management system, including the baggage trajectory [x, y, z, α, β, η], priority flag f, and system status code c. This data provides critical information for the control module, such as accurately locating the grab position based on the baggage trajectory, determining the processing order based on the priority flag, and flexibly adjusting the operating status based on the system status code. The control module also provides feedback on the loading status s to the baggage management system. If s = 0, a system-level error correction process is triggered to ensure accurate baggage handling.

[0049] Task Allocation: The baggage management system allocates tasks based on acquired baggage information and the real-time status of the robotic arms (busy or idle). High-priority bags are assigned to idle or efficient robotic arms to ensure rapid processing. When multiple bags are handled by multiple robotic arms, the system optimizes task scheduling and improves overall processing efficiency by taking into account factors such as the load, location, and destination of each robotic arm. For example, bags bound for the same flight can be allocated to a specific robotic arm, reducing the distance and time the robotic arm travels.

[0050] Dynamically adjust joint stiffness and degrees of freedom: The control module adjusts joint stiffness and degrees of freedom based on baggage information and system status provided by the baggage management system. When grasping irregularly shaped or loosely positioned luggage, the robot maintains high degrees of freedom and low joint stiffness, enhancing its flexibility and adapting to complex grasping requirements. During the handling and placement phases, the robot switches to a low degree of freedom and high joint stiffness, improving precision and stability and ensuring accurate placement of luggage.

[0051] System feedback and error correction: The robotic arm provides real-time feedback on loading status to the baggage management system. If a loading failure occurs (s=0), the baggage management system quickly initiates an error correction process. First, the robotic arm is instructed to perform a local retry, replanning the grasping path and adjusting grasping parameters. If the local retry fails, the system coordinates with other robotic arms or conveyor belts to perform a coordinated error correction, such as dispatching adjacent robotic arms for assistance or adjusting conveyor belt operation. If the problem persists, manual intervention is triggered, notifying staff to address the issue and ensure continued stable system operation.

[0052] The present invention interacts with the baggage handling system and intermediate management system through an interface module. The distributed architecture of ROS2 allows the robotic arm to be integrated into the airport's existing baggage handling system and intermediate management system as an independent module, eliminating the need for large-scale infrastructure modifications (such as conveyor belts and security equipment). This reduces deployment costs and allows for seamless integration with existing baggage management infrastructure.

[0053] The present invention also discloses a method for automatically sorting airport baggage using the above-mentioned airport baggage automatic sorting system based on the adaptive degree of freedom robotic arm, which comprises the following steps: S1, Baggage Transfer Stage: When a bag is transferred into the sensor module's collection range, the RFID reader acquires the baggage information, and the visual camera captures the baggage's precise position and posture. The collected data is then sent to the control module, which calculates the optimal grasping path based on the received data and configures the robot arm's degrees of freedom by adjusting the joint stiffness. S2, during the baggage grabbing phase, the robotic arm maintains a high degree of freedom configuration to ensure flexibility and grabs the baggage according to the optimal grabbing path calculated by the control module; During the baggage handling and placement phase, the robotic arm grabs the bag and switches to a low-degree-of-freedom configuration to improve accuracy and stability, moving and placing the baggage along the optimal path. During the entire process, the FBG sensor continuously monitors the strain state of the robotic arm body and provides real-time feedback for the control algorithm.

[0054] The value of this invention lies in three key aspects: First, through dynamic degree of freedom adjustment, the system can quickly adapt to baggage of varying sizes and shapes, resolving the problem of grasping failures caused by the fixed structure of traditional robotic arms; second, the intelligent control strategy dynamically adjusts the processing order based on baggage priority and system load, optimizing overall throughput; and finally, the system's seamless integration with existing baggage management infrastructure significantly reduces deployment costs while ensuring system reliability and maintainability.

[0055] The above embodiments and drawings do not limit the product form and style of the present invention. Any appropriate changes or modifications made by ordinary technicians in the relevant technical field should be deemed to be within the patent scope of the present invention.

Claims

1. An automatic airport baggage sorting system based on an adaptive degree of freedom robotic arm, characterized in that: include: Robotic arm body, sensor module, control module and interface module; The robotic arm body adopts a modular design, comprising multiple joints with adjustable stiffness and multiple connecting rods, each joint being connected to a control module; The sensing module includes an FBG sensor, a visual camera, and an RFID reader. The FBG sensor is embedded in the robotic arm body and connected to the control module. The visual camera is used to obtain luggage posture information, including the location and posture information of the luggage; the RFID reader is used to identify the luggage tag. The FBG sensor, visual camera, and RFID reader are respectively connected to the control module. The control module runs a reinforcement learning algorithm based on the Transformer architecture, dynamically adjusts the processing order based on baggage priority and system load, and outputs joint stiffness commands and corresponding drive commands to the robotic arm body; The interface module is integrated into the robotic arm body. The connector module exchanges data with the airport's baggage management system via the ROS 2 middleware connected to the control module. The data includes input information, including baggage trajectory, baggage priority, and system status, and output information, including loading status feedback.

2. The airport baggage automatic sorting system based on the adaptive degree of freedom robotic arm according to claim 1 is characterized by: The joint adopts a pneumatic-electromagnetic hybrid drive mode, and the stiffness of the joint Determined by the following formula: = × + × ； in, is the air pressure value of the jth joint, ranging from 0 to 1 MPa; is the electromagnetic coil current of the jth joint, ranging from 0 to 5A; is the aerodynamic stiffness coefficient, which depends on the diaphragm material and structure of the joint; is the electromagnetic stiffness coefficient, which is related to the performance of the permanent magnet and the number of coil turns; When the knuckles are locked, the critical value of the electromagnetic coil current; when ≥ , the joint enters a locked state and the robot's degrees of freedom are reduced.

3. The airport baggage automatic sorting system based on the adaptive degree of freedom robotic arm according to claim 2 is characterized by: The policy function π of the reinforcement learning algorithm is defined as =π( ), in, is the state observation at time t, including the luggage posture, conveyor belt speed v, and the joint angles and strain data of the robot arm; is the action output, including the and Set value; θ is the policy network parameter, obtained by minimizing the loss function ζ; θ= argmin g( , ); Argmin represents the set of independent variable values ​​that makes the function achieve the minimum value; Indicates that the discount factor is 0.99; The loss function ζ comprehensively considers task completion time, energy consumption, and operation accuracy: g = + ‖ ‖ The weight coefficient , , Determined through domain knowledge, Refers to the position of the luggage, Refers to the target posture of the luggage.

4. The airport baggage automatic sorting system based on the adaptive degree of freedom robotic arm according to claim 1 is characterized by: The input information of the interface module includes the baggage trajectory =[x,y,z,α,β,η]; x,y,z represent the three-dimensional coordinates of the luggage on the conveyor belt, α is the yaw angle of the luggage, β is the pitch angle of the luggage, and η is the roll angle of the luggage; Priority flag f∈{0,1}; when f=0, it indicates low-priority baggage, and when f=1, it indicates high-priority baggage; System status code c∈{0,...,255}; a binary numerical identifier indicating the overall operating status of the airport baggage handling system, with a value range of 0 to 255; The output information is the loading status feedback s∈{0,1}, and when s=0, the system-level error correction process is triggered.

5. The airport baggage automatic sorting system based on the adaptive degree of freedom robotic arm according to claim 1 is characterized by: The robotic arm body includes 6 joints with adjustable stiffness, and the connecting rod is made of carbon fiber reinforced polymer.

6. The airport baggage automatic sorting system based on the adaptive degree of freedom robotic arm according to claim 5 is characterized by: The FBG sensor in the sensing module is embedded in the carbon fiber reinforced polymer connecting rod of the robotic arm body, tightly integrated with the connecting rod and strained synchronously with the deformation of the connecting rod. The FBG sensor monitors the status of the robotic arm body in real time. By detecting the strain data of the connecting rod, it indirectly obtains the force conditions, motion posture and potential vibration risks of each joint of the robotic arm body, providing closed-loop feedback for the control algorithm. When the robotic arm grasps the body to pick up luggage of different weights or shapes, the FBG sensor provides real-time feedback on the connecting rod strain, assisting the control module in determining whether the joint stiffness needs to be adjusted, thereby achieving dynamic adjustment of the degrees of freedom.

7. The airport baggage automatic sorting system based on the adaptive degree of freedom robotic arm according to claim 5 is characterized by: The baggage management system and ROS 2 middleware implement data interaction through a collaborative optimization protocol, which dynamically allocates sorting tasks based on task priority and system load.

8. The airport baggage automatic sorting system based on the adaptive degree of freedom robotic arm according to claim 1 is characterized by: The robotic arm body, sensor module, control module and interface module are connected via wired or wireless means to achieve real-time data transmission and execution of control instructions.

9. The airport baggage automatic sorting method according to any one of claims 1 to 8, characterized in that: The following steps are involved: S1, Baggage Transfer Stage: When a bag is transferred into the sensor module's collection range, the RFID reader acquires the baggage information, and the visual camera captures the baggage's precise position and posture. The collected data is then sent to the control module, which calculates the optimal grasping path based on the received data and configures the robot arm's degrees of freedom by adjusting the joint stiffness. S2, during the baggage grabbing phase, the robotic arm maintains a high degree of freedom configuration to ensure flexibility and grabs the baggage according to the optimal grabbing path calculated by the control module; During the baggage handling and placement phase, the robotic arm grabs the bag and switches to a low-degree-of-freedom configuration to improve accuracy and stability, moving and placing the baggage along the optimal path. During the entire process, the FBG sensor continuously monitors the strain state of the robotic arm body and provides real-time feedback for the control algorithm.

10. The automatic airport baggage sorting method according to claim 9, characterized in that: When faced with luggage of different weights and shapes, the robotic arm automatically adjusts the air pressure and current values ​​of the joints according to the guidance of the strategy function.

Images