Object grasping method, system and robot based on robot buffering

Abstract

The application discloses a kind of object grabbing method, system and robot based on robot buffer, the method includes: real-time monitoring the motion trajectory of object thrown in air, and the flight state of object thrown in air is estimated and predicted by dynamic model and filtering algorithm;According to the flight state of object thrown in air, the coordination control between robot chassis, double arms and end effector is used to grab object thrown in air;Combined with end six-dimensional force sensor and admittance system, simulate spring damping in multiple directions, absorb the instantaneous impact force generated when object thrown in air and robot contact during grabbing object thrown in air.The application can accurately predict the trajectory and dynamic behavior of the object by estimating and predicting the flight state of the object, so that the robot can be adjusted in time before the object reaches the predetermined grabbing position, and the instantaneous impact force generated when the object and the robot contact is effectively relieved by designing spring damping, which greatly improves the safety of the robot grabbing process.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of robot grasping, and in particular to an object grasping method and system based on robot buffering, a robot and a computer readable storage medium. BACKGROUND

[0002] In modern industry, automated unloading has become one of the key technologies to improve efficiency and reduce labor costs. Traditional industrial unloading operations usually rely on manual or fixed mechanical equipment, and with the development of robot technology, more and more robots are introduced into unloading operations, especially in environments that require flexible handling of complex goods. In recent years, especially in scenarios such as warehousing, logistics and loading and unloading, the situation of fast moving or free falling objects is very common, which brings great challenges to the robot to capture and stabilize the objects. Therefore, how to realize the mobile robot to capture and accurately grasp the free falling objects has become a hot topic in the field of intelligent industrial unloading.

[0003] The existing robot unloading has the following shortcomings:

[0004] Lack of real-time and adaptability: In the actual scenario of industrial unloading, the landing point, speed and direction of the object may have great variability, which makes the existing technology unable to quickly adapt to complex and uncertain environments, and does not consider the buffering of objects with large mass and inertia.

[0005] Fixed platform limitation: Many dual-arm robots or industrial robots rely on fixed platforms, which limits their ability to quickly react and adjust in complex dynamic scenarios. For tasks that require flexible movement, such as grasping falling objects, fixed platforms cannot meet the task requirements and cannot make effective real-time adjustments according to the object motion trajectory.

[0006] Lack of multi-task collaboration: Although existing dual-arm robots can perform multi-task collaboration, they often do not consider the coordination and cooperation between the mobile platform and the dual arms. In scenarios that require rapid position adjustment and object capture, there is a lack of effective collaborative control between the chassis and the dual arms, which prevents the robot from quickly and accurately capturing and stabilizing the object.

[0007] Therefore, the existing technology still needs to be improved and developed. SUMMARY

[0008] The main purpose of the present application is to provide an object grasping method and system based on robot buffering, a robot and a computer readable storage medium, which aims to solve the problem of low safety in the robot grasping process in the prior art, and the problem of inability to smoothly and efficiently grasp objects, and the problem of hardware damage to the robot.

[0009] To achieve the above object, the application provides a robot buffer-based object grabbing method, which comprises the following steps:

[0010] Real-time monitoring of the motion trajectory of the air-throwing object, estimation and prediction of the flight state of the air-throwing object through a dynamic model and a filtering algorithm;

[0011] According to the flight state of the air-throwing object, coordinated control among the robot chassis, the double arms and the end effector is utilized to grab the air-throwing object;

[0012] Combination of the end six-dimensional force sensor and the admittance system to simulate spring damping in multiple directions, absorption of the instantaneous impact force generated when the air-throwing object contacts the robot during the grabbing of the air-throwing object.

[0013] Optionally, the robot buffer-based object grabbing method, wherein the real-time monitoring of the motion trajectory of the air-throwing object and the estimation and prediction of the flight state of the air-throwing object through a dynamic model and a filtering algorithm are specifically as follows:

[0014] Adoption of a dynamic capture system to capture the reflective markers attached to the air-throwing object to capture the object information, combination of a parabolic state equation to estimate the motion velocity of the air-throwing object and predict the motion trajectory of the air-throwing object.

[0015] Optionally, the robot buffer-based object grabbing method, wherein the estimation of the motion velocity of the air-throwing object and the prediction of the motion trajectory of the air-throwing object are specifically calculated as follows:

[0016]

[0017] Wherein, k represents the current time, represents the object state estimated at the last time, F k represents the motion matrix, represents the object state estimated at the last time, B k represents the gravity matrix, u k represents the gravity, represents the estimated current object state, K k represents the Kalman estimation, z k represents the observation state, H k represents the observation matrix, P k|k-1 represents the covariance matrix, P k-1|k-1 represents the prior covariance matrix at the last time, T represents the transpose, Q k represents the state noise matrix, R k represents the observation noise matrix, t represents the estimation time, x0 represents the initial position of the x-axis in the world Cartesian coordinate system, y0 represents the initial position of the y-axis in the world Cartesian coordinate system, z0 represents the initial position of the z-axis in the world Cartesian coordinate system, υ0x represents the initial speed of the x-axis in the world Cartesian coordinate system, υ 0y represents the initial speed of the y-axis in the world Cartesian coordinate system, υ 0z represents the initial speed of the z-axis in the world Cartesian coordinate system, x(t) represents the position of the x-axis in the world Cartesian coordinate system, y(t) represents the position of the y-axis in the world Cartesian coordinate system, z(t) represents the position of the z-axis in the world Cartesian coordinate system, and g represents the acceleration of gravity.

[0018] Optionally, the object grabbing method based on robot buffering, wherein the grabbing of the flying object according to the flight state of the flying object by using the coordinated control among the robot chassis, the dual arms and the end effector, specifically comprises:

[0019] Based on the dynamic characteristics of the robot itself, an interaction model of the robot and the flying object is established, and an optimal grabbing trajectory with the shortest time is calculated.

[0020] The whole-body model predictive control technology is adopted to coordinate the movements of the chassis, the mechanical arms and the end effector, and to control the robot to accurately track the movement trajectory of the flying object for accurate grabbing.

[0021] Optionally, the object grabbing method based on robot buffering, wherein the coordinated movement of the chassis, the mechanical arms and the end effector specifically comprises:

[0022] According to the kinematic formula of the chassis and the kinematic equation of the mechanical arms, the joint angular velocity and the chassis movement speed are optimized and solved.

[0023] Optionally, the object grabbing method based on robot buffering, wherein the combination of the end six-dimensional force sensor and the admittance system simulates spring dampers in multiple directions to absorb the instantaneous impact force generated when the flying object contacts the robot during the grabbing of the flying object, specifically comprises:

[0024] The combination of the end six-dimensional force sensor and the admittance system simulates spring dampers in multiple directions to obtain additional simulated forces for buffering, and the additional simulated forces are used to relieve the impact force caused by the flying object to the robot during the grabbing.

[0025] Optionally, the object grabbing method based on robot buffering, wherein the calculation of the additional simulated forces specifically comprises:

[0026]

[0027] wherein M represents an inertia coefficient, represents the end displacement acceleration, and B represents a damping coefficient, wherein K represents the elastic coefficient, Δx represents the end displacement, X represents the end position, and F represents the additional simulated force.

[0028] In addition, the application also provides a robot buffer-based object grabbing system, which comprises:

[0029] a real-time prediction module for monitoring the motion trajectory of the air-throwing object in real time and estimating and predicting the flight state of the air-throwing object through a dynamic model and a filtering algorithm;

[0030] a full-body control module for grabbing the air-throwing object by coordinating the control among the robot chassis, the double arms and the end effector according to the flight state of the air-throwing object;

[0031] a full-body buffer module for simulating spring dampings in multiple directions in combination with the end six-dimensional force sensor and the admittance system to absorb the instantaneous impact force generated when the air-throwing object contacts the robot during the grabbing process.

[0032] In addition, the application also provides a robot, which comprises a memory, a processor and a robot buffer-based object grabbing program stored in the memory and executable on the processor, and the robot buffer-based object grabbing program realizes the steps of the robot buffer-based object grabbing method as described above when executed by the processor.

[0033] In addition, the application also provides a computer readable storage medium, which stores a robot buffer-based object grabbing program, and the robot buffer-based object grabbing program realizes the steps of the robot buffer-based object grabbing method as described above when executed by a processor.

[0034] In the application, the motion trajectory of the air-throwing object is monitored in real time, and the flight state of the air-throwing object is estimated and predicted through a dynamic model and a filtering algorithm; the air-throwing object is grabbed by coordinating the control among the robot chassis, the double arms and the end effector according to the flight state of the air-throwing object; and the spring dampings in multiple directions are simulated in combination with the end six-dimensional force sensor and the admittance system to absorb the instantaneous impact force generated when the air-throwing object contacts the robot during the grabbing process. The application can accurately predict the trajectory and dynamic behavior of the object by estimating and predicting the flight state of the object, so that the robot can be adjusted in time before the object reaches the predetermined grabbing position, the instantaneous impact force generated when the object contacts the robot is effectively relieved by designing the spring damping system or the energy-absorbing material, the safety during the grabbing process of the robot is greatly improved, the robot can stably and efficiently grab the object, and the loss of the robot hardware is reduced. Attached Figure Description

[0035] Figure 1 This is a flowchart of a preferred embodiment of the object grasping method based on robot buffering of the present invention;

[0036] Figure 2 This is a schematic diagram of three processes of a robot grasping an object thrown from the air in a preferred embodiment of the robot-buffered object grasping method of the present invention.

[0037] Figure 3 This is a schematic diagram illustrating the principle of object grasping based on dual-arm collaborative control in a preferred embodiment of the robot-buffered object grasping method of the present invention.

[0038] Figure 4 This is a structural diagram of a preferred embodiment of the object grasping system based on robot buffering of the present invention;

[0039] Figure 5 This is a structural diagram of a preferred embodiment of the robot of the present invention. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0041] The preferred embodiment of the object grasping method based on robot buffering described in this invention, such as... Figure 1 , Figure 2 and Figure 3 As shown, the object grasping method based on robot buffering includes the following steps:

[0042] Step S10: Monitor the trajectory of the air-dropped object in real time, and estimate and predict the flight state of the air-dropped object through dynamic models and filtering algorithms.

[0043] Specifically, such as Figure 2 As shown, the object state is estimated and predicted: a dynamic capture system is used to capture reflective marks attached to the air-launched object to obtain object information. Using the six-bit attitude obtained from the object information, a Kalman filter combined with the parabolic state equation is used to estimate the velocity and predict the trajectory of the air-launched object. Through dynamic models and filtering algorithms, the flight state of the object can be accurately estimated and predicted, providing precise spatiotemporal information for subsequent grasping actions. The specific calculations are as follows:

[0044]

[0045] Where k represents the current time, F represents the estimated state of the object at the previous moment.k represents a motion matrix, represents the state of the object estimated at the last time, B k represents a gravity matrix, u k represents gravity, represents the state of the object estimated at the last time, B k represents a Kalman estimation, z k represents an observation state, H k represents an observation matrix, P k|k-1 represents a covariance matrix, P k-1|k-1 represents a prior covariance matrix at the last time, T represents transposition, Q k represents a state noise matrix, R k represents an observation noise matrix, t represents an estimation time, x0 represents an initial position of an x axis in a world Cartesian coordinate system, y0 represents an initial position of a y axis in the world Cartesian coordinate system, z0 represents an initial position of a z axis in the world Cartesian coordinate system, υ 0x represents an initial velocity of the x axis in the world Cartesian coordinate system, υ 0y represents an initial velocity of the y axis in the world Cartesian coordinate system, υ 0z represents an initial velocity of the z axis in the world Cartesian coordinate system, x(t) represents a position of the x axis in the world Cartesian coordinate system, y(t) represents a position of the y axis in the world Cartesian coordinate system, z(t) represents a position of the z axis in the world Cartesian coordinate system, g represents a gravity acceleration.

[0046] In step S20, the flying state of the empty thrown object is used to control the coordination between the robot chassis, the two arms and the end effector to grasp the empty thrown object.

[0047] Specifically, as shown in FIG. 4, the grasping optimization is performed as follows. Figure 2 Based on the dynamic characteristics of the robot itself, an interaction model of the robot and the empty thrown object is established, and an optimal grasping trajectory with the shortest time is calculated. The full-body model predictive control technology is used to coordinate the movements of the chassis, the mechanical arms and the end effector, and to control the robot to accurately track the movement trajectory of the empty thrown object for precise grasping.

[0048] The robot not only relies on the grasping action of the upper limbs, but also combines the movement of the chassis to realize the collaborative work of the robot chassis, the two arms and the end effector through coordinated control. After the desired end position (Xref) is input, the robot system will optimize and solve the joint angular velocity and the chassis movement speed according to the kinematic formula of the two chassis and the kinematic equation of the mechanical arm, and the weight of the movement is determined by the energy consumption ratio (R matrix). Each joint and chassis direction has an energy weight item, and the higher the weight of the mechanical arm joint, the more the chassis will be moved, and the higher the weight of the chassis, the more the mechanical arm will be controlled.

[0049] In the process of grabbing the flying object, the application considers the dynamic characteristics of the robot itself, establishes an interaction model of the robot and the thrown object (considering the end control position, converting the expected position into force to buffer the impulse of the robot catching the object), calculates the optimal grabbing trajectory with the shortest time (considering the kinematics equation of the end and the kinematics equation of the object, and inputting the landing position and time of the object as the minimum constraint for optimization), adopts the whole body model predictive control technology (considering the whole body model predictive control, not only considering the current state, and combining the kinematics equation of the robot to predict the next control, so that the system converges faster and is more robust), so that the robot can accurately track the motion trajectory of the object and realize accurate grabbing.

[0050] Step S30, combined with the end six-dimensional force sensor and the admittance system, simulate spring damping in multiple directions, absorb the instantaneous impact force generated when the thrown object contacts the robot during the process of grabbing the thrown object.

[0051] Specifically, as shown in Figure 2 Impulse buffering: combined with the end six-dimensional force sensor and the admittance system, simulate spring damping in multiple directions, get additional simulated forces for buffering, use the additional simulated forces to alleviate the impact force caused by the thrown object on the robot during the grabbing process, can effectively absorb the instantaneous impact force generated when the object contacts the robot, ensure that the object is gently caught by the robot, thereby avoiding damage to the robot caused by collision.

[0052] Wherein, the calculation of the additional simulated force is:

[0053]

[0054] Wherein, M represents the inertia coefficient, represents the end displacement acceleration, B represents the damping coefficient, represents the end displacement velocity, K represents the elastic coefficient, △x represents the end displacement, X represents the end position, and F represents the additional simulated force.

[0055] The technical scheme of the application not only realizes accurate grabbing of the flying object by the robot, but also significantly improves the stability and flexibility of the robot system. By predicting the flight trajectory of the object in real time, the robot can accurately capture the flying object and complete the grabbing action with the optimal trajectory. By coordinating the movement of the chassis and the mechanical arm, the accuracy and efficiency of grabbing are improved, especially in a dynamic environment, the balance and stability of the system can be ensured. In addition, the buffer effectively reduces the impact force in the grabbing process, protects the safety of the robot body, and prolongs the service life of the equipment. The elastic design of the system ensures the soft contact with the object during the grabbing process, avoids the damage of the grabbed object, and further improves the success rate and reliability of the grabbing.

[0056] The present application not only solves the precision and flexibility problems in traditional robot grasping tasks, but also significantly improves the adaptability and work efficiency of the robot. Through this cooperative control mode, the robot can perform more accurate grasping operations in complex environments, especially when facing dynamic objects and complex movements, showing higher flexibility and response ability.

[0057] In addition, the present application also innovatively realizes impact force buffering and ground adaptability. Through the cooperative work of the chassis and the upper limbs, combined with the whole body buffering, the robot can effectively relieve the impact force generated when contacting the object. The movement of the chassis can help the robot absorb and disperse the impact force during grasping, avoiding excessive impact that causes the robot to be unstable or damaged. This innovative technology enables the robot to maintain stability and protect its structure and function when grasping heavy or high-speed objects.

[0058] In addition, the current design mainly focuses on single dual-arm robot for grasping tasks, and in the future, multiple robots can be considered to work cooperatively to improve the unloading efficiency through a networked multi-robot system. Under this scheme, multiple robots can be assigned tasks and cooperatively grasp in real time, further improving the efficiency and precision of the overall system.

[0059] Further, as shown in Figure 4 Based on the above robot buffering-based object grasping method, the present application also correspondingly provides a robot buffering-based object grasping system, wherein the robot buffering-based object grasping system comprises:

[0060] A real-time prediction module 51 is used to monitor the motion trajectory of the thrown object in real time, and estimate and predict the flight state of the thrown object through a dynamic model and a filtering algorithm;

[0061] A whole body control module 52 is used to grasp the thrown object according to the flight state of the thrown object, using coordinated control between the robot chassis, dual arms and end effector;

[0062] A whole body buffering module 53 is used to simulate multiple direction spring dampers in combination with the end six-axis force sensor and the admittance system, to absorb the instantaneous impact force generated when the thrown object contacts the robot during grasping.

[0063] In the present application, the real-time prediction module 51 monitors the motion trajectory of the object in real time, and estimates and predicts the flight state of the object through a dynamic model and a filtering algorithm; the whole body control module 52 utilizes the coordinated control between the robot chassis, the double arms and the end effector, and in the process of grabbing the empty thrown object, the robot not only relies on the grabbing action of the upper limbs, but also combines the movement of the chassis to ensure the balance and stability of the whole system; the whole body buffer module 53 is used to relieve the impact force caused by the object to the robot during the grabbing process, and through the design of the spring damping system, the instantaneous impact force generated when the object contacts the robot can be effectively absorbed.

[0064] As shown in Figure 3 , the robot system first estimates the physical state of the object and predicts its flight trajectory, and then transmits the estimated state to the whole body control module; the planner simplifies the joint calculation of the robot, helps the trajectory of the object grabbing end, and transmits it to the whole body control module (MPC), and the controller then outputs the joint speed of each joint. These speed signals are converted into torque instructions for the underlying motor through the PD controller.

[0065] Technical effects of the present application:

[0066] (1) Improve the capture accuracy and success rate: the present application can accurately predict the trajectory and dynamic behavior of the object through the real-time object motion prediction module combined with advanced algorithms such as extended Kalman filter, so that the robot can make timely adjustments before the object reaches the predetermined grabbing position.

[0067] (2) Expand the motion space and flexibility: the whole body coordinated control scheme (whole body control module) of the present application effectively integrates the motion of the robot chassis and the double arms, so that the robot can fully utilize the movement of the chassis to adjust the position and posture of the upper limbs when performing the grabbing task.

[0068] (3) Impact force buffering and shock absorption: the whole body buffer module of the present application effectively relieves the instantaneous impact force generated when the object contacts the robot by designing a spring damping system or energy-absorbing material. Through the response of the whole body buffer module, the kinetic energy of the object can be effectively absorbed, avoiding the damage or unstable operation of the robot caused by excessive impact force.

[0069] The present application greatly improves the safety of the robot during the grabbing process, ensures that the robot can smoothly and efficiently grab the object, and reduces the wear and tear of the robot hardware.

[0070] Further, as shown in Figure 5 , based on the above object grabbing method and system based on robot buffering, the present application also correspondingly provides a robot, which comprises a processor 10, a memory 20 and a display 30. Figure 5Only some components of the robot are shown, but it should be understood that all of the components shown are not required, and that more or fewer components can be substituted.

[0071] The memory 20 can be an internal storage unit of the robot, such as a hard disk or a memory of the robot, in some embodiments. The memory 20 can also be an external storage device of the robot, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc., equipped on the robot, in other embodiments. Further, the memory 20 can include both an internal storage unit and an external storage device of the robot. The memory 20 is used to store application software installed on the robot and various types of data, such as program codes of the installed robot, etc. The memory 20 can also be used to temporarily store data that has been output or will be output. In an embodiment, a robot buffer-based object grasping program 40 is stored on the memory 20, which can be executed by the processor 10 to implement the robot buffer-based object grasping method described in the present application.

[0072] The processor 10 can be a Central Processing Unit (CPU), a microprocessor, or other data processing chip, in some embodiments, which is used to run program codes or process data stored in the memory 20, such as to execute the robot buffer-based object grasping method, etc.

[0073] The display 30 can be an LED display, a liquid crystal display, a touch liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, etc., in some embodiments. The display 30 is used to display information of the robot and to display a visualized user interface. The processor 10, the memory 20, and the display 30 of the robot communicate with each other through a system bus.

[0074] In an embodiment, the steps of the robot buffer-based object grasping method described above are implemented when the processor 10 executes the robot buffer-based object grasping program 40 in the memory 20.

[0075] The present application also provides a computer readable storage medium, wherein the computer readable storage medium stores a robot buffer-based object grasping program, which, when executed by a processor, implements the steps of the robot buffer-based object grasping method described above.

[0076] In summary, the application provides an object grabbing method, system, robot and computer readable storage medium based on robot buffering, the method comprising: monitoring the motion trajectory of the object thrown in the air in real time, and estimating and predicting the flight state of the object thrown in the air through a dynamic model and a filtering algorithm; according to the flight state of the object thrown in the air, grabbing the object thrown in the air by coordinated control among the robot chassis, the double arms and the end effector; combining the end six-dimensional force sensor and the admittance system, simulating spring damping in multiple directions, and absorbing the instantaneous impact force generated when the object thrown in the air contacts the robot during the process of grabbing the object thrown in the air. The application can accurately predict the trajectory and dynamic behavior of the object by estimating and predicting the flight state of the object, so that the robot can be adjusted in time before the object reaches the predetermined grabbing position, the instantaneous impact force generated when the object contacts the robot is effectively relieved by designing a spring damping system or an energy absorption material, the safety of the robot during the grabbing process is greatly improved, the robot can stably and efficiently grab the object, and the loss of the robot hardware is reduced.

[0077] It should be noted that in this paper, the term "include", "contain" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or robot including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or includes elements inherent to such process, method, article or robot. Without more limitations, the element defined by the sentence "includes a" does not exclude the existence of another same element in the process, method, article or robot including the element.

[0078] Of course, those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing relevant hardware (such as a processor, a controller, etc.) through a computer program, and the program can be stored in a computer readable computer readable storage medium, and the program can include the processes of the above-mentioned method embodiments when executed. The computer readable storage medium can be a memory, a magnetic disc, an optical disc, etc.

[0079] It should be understood that the application is not limited to the above examples, and those skilled in the art can improve or change it according to the above description, and all these improvements and changes should belong to the protection scope of the claims of the application.

Claims

1. A method for object grasping based on robot buffering, characterized in that, The object grasping method based on robot buffering includes: The trajectory of objects thrown from the air is monitored in real time, and the flight state of the objects is estimated and predicted through dynamic models and filtering algorithms. Based on the flight state of the object thrown from the air, the robot uses coordinated control between its chassis, arms, and end effector to grab the object. By combining a six-dimensional force sensor and admittance system at the end, and simulating spring damping in multiple directions, the robot absorbs the instantaneous impact force generated when the object comes into contact with the robot during the grasping process. The combination of an end-effector six-dimensional force sensor and admittance system simulates multi-directional spring damping, absorbing the instantaneous impact force generated when the object comes into contact with the robot during the grasping process. Specifically: By combining a six-dimensional force sensor at the end and an admittance system, spring damping in multiple directions is simulated to obtain additional simulated forces for buffering. These additional simulated forces are used to mitigate the impact of objects thrown from the air onto the robot during the grasping process. The calculation of the additional simulated force is as follows: in, Indicates the coefficient of inertia. Indicates the terminal displacement acceleration. Indicates the damping coefficient. Indicates the end displacement velocity. Represents the elastic coefficient. Indicates the end displacement. Indicates the end position. This represents the additional simulated force.

2. The object grasping method based on robot buffering according to claim 1, characterized in that, The real-time monitoring of the trajectory of the air-launched object, and the estimation and prediction of the object's flight state through dynamic models and filtering algorithms, specifically includes: A dynamic capture system is used to capture reflective markings on objects thrown from the air to obtain object information. Combined with the parabolic state equation, the velocity of the thrown object is estimated and the trajectory of the thrown object is predicted.

3. The object grasping method based on robot buffering according to claim 2, characterized in that, The estimation of the velocity of the object thrown from the air and the prediction of its trajectory are specifically calculated as follows: ； ； ； ； in, k Indicates the current moment. This represents the estimated state of the object at the previous moment. Represents the motion matrix. This represents the estimated state of the object at the previous moment. Represents the gravity matrix. Represents gravity. This indicates an estimate of the current state of the object. This indicates that the Kalman estimate, Indicates the observation status. Represents the observation matrix. Represents the covariance matrix. Let the prior covariance matrix at the previous time step be denoted as . Indicates transpose. Represents the state noise matrix. Represents the observation noise matrix. Indicates estimated time. Represents the world Cartesian coordinate system x Initial position of the shaft Represents the world Cartesian coordinate system y Initial position of the shaft Represents the world Cartesian coordinate system z Initial position of the shaft Represents the world Cartesian coordinate system x Initial shaft velocity, Represents the world Cartesian coordinate system y Initial shaft velocity, Represents the world Cartesian coordinate system z Initial shaft velocity, Represents the world Cartesian coordinate system x Axis position, Represents the world Cartesian coordinate system y Axis position, Represents the world Cartesian coordinate system z Axis position, It represents the acceleration due to gravity.

4. The object grasping method based on robot buffering according to claim 1, characterized in that, The process of grasping the air-launched object by coordinating the robot chassis, arms, and end effector based on its flight trajectory includes: Based on the robot's own dynamic characteristics, an interaction model between the robot and the object thrown from the air is established, and the optimal grasping trajectory with the shortest time is calculated. Using whole-body model predictive control technology, the movement of the chassis, robotic arm and end effector is coordinated to control the robot to accurately track the trajectory of objects thrown from the air for precise grasping.

5. The object grasping method based on robot buffering according to claim 4, characterized in that, The coordination of the movement of the chassis, robotic arm, and end effector is specifically as follows: Based on the kinematic formulas of the chassis and the kinematic equations of the robotic arm, the joint angular velocity and chassis motion velocity are obtained through optimization.

6. A robot-based object grasping system, characterized in that, The robot-buffered object grasping system is used to implement the robot-buffered object grasping method according to any one of claims 1-5, wherein the robot-buffered object grasping system comprises: The real-time prediction module is used to monitor the trajectory of objects thrown from the air in real time, and to estimate and predict the flight state of the objects through dynamic models and filtering algorithms. The whole-body control module is used to grasp the air-dropped object by coordinating the control between the robot chassis, the two arms and the end effector according to the flight state of the object. The full-body buffer module, combined with the end effector six-dimensional force sensor and admittance system, simulates spring damping in multiple directions to absorb the instantaneous impact force generated when the object comes into contact with the robot during the grasping process.

7. A robot, characterized in that, The robot includes: a memory, a processor, and a robot-buffered object grasping program stored in the memory and executable on the processor, wherein the robot-buffered object grasping program, when executed by the processor, implements the steps of the robot-buffered object grasping method as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a robot-buffered object grasping program, which, when executed by a processor, implements the steps of the robot-buffered object grasping method as described in any one of claims 1-5.

Images