Adaptation of a gripping simulation by way of parameter identification in the real world
Patent Information
- Application Number
- EP2023793796
- Authority / Receiving Office
- EP · EP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2022-11-16
- Filing Date
- 2023-10-20
- Publication Date
- 2025-09-24
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Current simulation technologies face a simulation-to-reality gap, as they are often trained in environments with generated data that only qualitatively reflect real-world conditions, leading to suboptimal parameterization and reduced accuracy in robot-assisted gripping processes.
A method is developed to automatically optimize parameters for robot-assisted gripping by determining real gripping and holding data, which includes forces, friction coefficients, and object properties, and using this data to simulate and adjust parameters in a simulation environment based on a cost function, allowing for more realistic simulations and improved gripping success.
This approach reduces the simulation-to-reality gap by enabling more accurate simulation environments, enhancing gripping success rates and parameter optimization, particularly in scenarios where objects are gripped near edges or fall due to friction issues, and allows for quicker optimization of dynamic parameters.
Smart Images
Figure 1.1
Abstract
Description
[0001] Description
[0002] Adaptation of a gripping simulation through parameter identification in the real world
[0003] The present invention relates to a method for the automated optimization of parameters, in particular for a robot-assisted gripping process, a method for controlling a gripping robot, a system for the automated optimization of parameters, and a computer program or computer program product.
[0004] In the current state of the art, technical systems are sometimes trained in simulation environments. For this purpose, simulated data is typically generated in any quantity. However, this simulated data is only a qualitative representation of reality. The difference between simulation and reality is commonly referred to as the simulation-to-reality gap (Sim2Real gap).
[0005] The object of the present invention is in particular to reduce the simulation-to-reality gap and, in particular, to optimize the parameterization of the simulation.
[0006] This object is achieved according to the teaching of the independent claims. Various embodiments and further developments of the invention are the subject of the dependent claims.
[0007] In one embodiment of the present invention, a method is provided for automatically optimizing parameters, in particular for a robot-assisted gripping process. In one embodiment, the method comprises determining real gripping data, wherein the real gripping data describe at least one gripping success of a gripping position, in particular on an object to be gripped by a gripping robot or on an object gripped by a gripping robot. Alternatively or additionally, the method comprises determining real holding data, wherein the real holding data describe at least one force, in particular a force on the object gripped by a gripping robot and / or on the gripper of the gripping robot, which has in particular (successfully) gripped an object to be gripped. In one embodiment, the real holding data describe at least one parameter orrelevant property, in particular a parameter or property associated with the grip. In one embodiment, a relevant parameter or property is a force, in particular a normal force, a frictional force, in particular a friction or a friction coefficient, such as in particular (for) static friction, rolling friction, rolling friction, spinning friction, and / or lateral friction, which acts on the gripping object and / or on at least one gripper finger, in particular a closing force of the at least one gripper finger, and / or a mass, inertia and / or position of a center of mass of the gripping object and / or of the at least one gripper finger.In one embodiment, the method further comprises, in particular in one step, simulating, in particular replicating, further in particular in a simulation environment, wherein at least one gripping process, in particular a grip, is simulated in the simulation environment based on the gripping process, in particular a grip, underlying the determined real gripping data and / or the determined real holding data. In one embodiment, the method further comprises, in particular in one step, data-based optimization of parameters of the simulated gripping process, in particular of the simulated grip. Alternatively or additionally, the method comprises optimizing the parameters of the simulated gripping process, in particular of the simulated grip, based on the determined real gripping data and / or based on the determined real holding data.
[0008] In this way, in one embodiment, optimized parameters can advantageously be determined, in particular an improved simulation environment can advantageously be achieved, further in particular a more realistic or near-realistic / closer-to-realistic simulation or simulation environment, in particular in comparison with simulations or simulation environments that are not optimized or optimized based on real gripping data and / or real holding data.
[0009] The term "real gripping data," as used herein, is to be understood in particular as information, more particularly digital information, that describes the data of a real gripping process, in particular a grip actually performed. In particular, the term, in one embodiment, is to be understood as a counterpart to "gripping
[0010] Simulation data” can be understood, whereby the “gripping simulation data” refers to information that describes data of a simulated gripping process, in particular a simulated grip, in a simulation environment.
[0011] The term "real holding data," as used herein, should be understood in particular as information, more particularly digital information, describing the data of a holding action during or during an actual gripping action, in particular during or during an actual grip. In particular, the term, in one embodiment, should be understood as the counterpart to "holding simulation data."
[0012] In one embodiment, the “real data” and the “simulation data” each describe the same thing or information, respectively for the reality and the simulation, or the simulation environment.
[0013] In one embodiment, the data-based optimization comprises an automatic adjustment of the parameters, in particular an automatic adjustment of a simulated mass of the simulation, an automatic adjustment of (simulated) dynamic parameters and / or a simulated force, in particular friction force or friction. In one embodiment, the automatic adjustment of the parameters is based on at least one evaluation of a cost function that underlies the optimization or is optimized on the basis of.
[0014] In one embodiment, the data-based optimization, in particular the optimization of the parameters, is based on a cost function of real gripping data and gripping simulation data and / or real holding data and holding simulation data. In one embodiment, the gripping simulation data describe at least one simulated gripping parameter, in particular a gripping success of a gripping position in the simulation. In one embodiment, the holding simulation data describe at least one simulated holding parameter, in particular a force of the simulated grip.In one embodiment, the cost function can be based on a comparison of matching entries in vectors of successful grasps in reality and the simulation, in particular a cost function c = sum(amount(grasp_success_real - grasp_success_sim)), where grasp_success_real corresponds to a vector for a grasp success in reality, in particular for a grasp executed in real life, and grasp_success_sim corresponds to a vector for a grasp success in a corresponding grasp in the simulation environment. In one embodiment, the entries in the vectors can be 1 or 0, corresponding to a successful grasp or an unsuccessful grasp.In one embodiment, the data-based optimization comprises the adaptation of, in particular abstract, parameters, such as in particular time steps of the simulation, solver iterations or the like, further in particular "lateral friction", "rolling friction", "spinning friction" of the gripping object and / or "lateral friction", "rolling friction", "spinning friction" of the gripper, in particular of at least one gripper finger, a mass, an inertia and / or a center of mass of the gripping object, or the like.
[0015] This advantageously enables decisions regarding the parameters, in particular dynamic parameters, to be estimated based on evaluations of the cost function. Thus, in one embodiment, if a grip is located close to the edge of the object and / or at the fingertips of the gripper fingers, and if the object cannot be held or falls out of the gripper or slips out of the fingertips, the object can be made heavier in the simulation or a friction parameter can be adjusted, particularly advantageously automatically and / or data-based.
[0016] In one embodiment, the optimization, in particular the data-based optimization, is performed using a gradient-free optimizer. In one embodiment, the optimization algorithm is based on a Monte Carlo algorithm or the optimization algorithm is a Monte Carlo algorithm, in particular a covariance matrix adaptation evolution strategy (CMA-ES) or the like.
[0017] This can, in one embodiment, enable the optimization of parameters to be carried out more quickly. Advantageously, this can be done automatically in one embodiment.
[0018] In one embodiment, determining real gripping data comprises determining a plurality of, in particular collision-free and / or graspable, gripping positions on the (real) object to be grasped. In one embodiment, determining real gripping data further comprises determining a plurality of, in particular collision-free and / or graspable, gripping positions. In one embodiment, determining real gripping data further comprises selecting one of the determined gripping positions. In one embodiment, determining real gripping data further comprises gripping the object, in particular with a gripping robot, based on the selected gripping position. In one embodiment, determining real gripping data further comprises storing the gripping success for the selected gripping position, in particular for the selected gripping position on the object to be grasped and for a pose of the object to be grasped, in particular on a work surface of the gripping robot.
[0019] In one embodiment, the determination of real gripping data comprises a step, in particular a preceding step, with randomized placement of an object or objects to be gripped, in particular with the gripping robot.
[0020] In one embodiment, determining gripping real data further comprises repeating the method steps described herein for determining gripping real data.
[0021] In one embodiment, the object to be gripped is moved by the robot, in particular after determining real gripping data, to a random position, in particular to a random position in the working area of the gripping robot, and then dropped.
[0022] In some embodiments, this allows the object to be grasped to land in random poses, particularly on a work surface of the robot. Accordingly, in some embodiments, grip positions are determined that correspond to the random pose of the object to be grasped.
[0023] Advantageously, in some embodiments, a pre-sorted number of grips for the object can be obtained, in particular according to gripping success.
[0024] In one embodiment of the present invention, a method for optimizing a grip sampler is provided. In one embodiment, the method comprises, at least substantially, the same steps as determining real grip data, in particular with a selection of one of the determined grip positions based on a selection frequency of the grip positions, as described herein. Advantageously, in one embodiment, a grip sampler can be optimized using the determined real grip data, in particular a grip with a high or higher grip success can be determined more quickly, in particular by the grip sampler, further in particular based on the determined, in particular stored, real grip data.
[0025] In one embodiment, the selection of one of the determined grip positions, in particular when repeating the method steps described herein for determining real grip data, is based on a selection frequency of the determined grip positions, in particular the determined grip position with the lowest selection frequency of the determined grip positions is selected.
[0026] In this way, a pre-sorted set of grip positions for an object to be grasped can advantageously be obtained, in particular more quickly.
[0027] In one embodiment, determining real holding data comprises determining a holding grip position, in particular on a gripped object, wherein the holding grip position is arranged on the object to be gripped and / or on the gripped object in such a way that a force can be exerted on the object to be gripped and / or on the gripped object via the holding grip position, which force can act or acts opposite to at least one force on the object to be gripped and / or on the gripped object, in particular opposite to a gripping direction of the gripping robot. In one embodiment, determining real holding data further comprises gripping, in particular holding, the object at the determined holding grip position, in particular by a holding robot different from the gripping robot.In one embodiment, this can comprise compensating for the gravity acting on the gripped object, particularly when the object is large compared to the gripper or gripping robot, and / or heavy, particularly compared to the maximum force of the gripping robot. In one embodiment, determining real holding data further comprises exerting a force, particularly via the holding grip position, furthermore in particular by means of the holding robot, on the object, in particular a force that acts oppositely to a force on the gripped object, furthermore in particular a force that acts oppositely to a force exerted on the object by the gripping robot (not by the holding robot), in particular oppositely to a gripping direction of the gripping robot. In one embodiment, this can advantageously make it possible for parameter values for a simulation to be determined more precisely, particularly quickly.can be determined. Advantageously, in one embodiment, this makes it possible for rotational forces, in particular rotational friction coefficients, to be determined better than without the use of a holding robot or based solely on a simulation environment.
[0028] In one embodiment, the application of the force, in particular by means of the holding robot, involves a successive increase of the force in predetermined steps, in particular until the gripping robot loses the grasped object and / or until a, in particular predetermined, maximum force is exceeded. In one embodiment, the application of the force involves a (numerical) measurement of the applied force.
[0029] This advantageously makes it possible, in one embodiment, to determine parameters for a simulation more precisely, particularly step by step. Furthermore, in one embodiment, it is advantageously possible to determine more precisely at what applied force the grip was lost.
[0030] In one embodiment, determining real gripping data and / or real holding data comprises capturing the object using a capturing device. In one embodiment, capturing the object comprises determining a pose of the object, in particular over time, and further comprises, in particular, localizing the grasped object, in particular in the gripper (in-hand localization), and / or tracking the object, in particular while it is gripped.
[0031] In one embodiment, in particular by recording the object with the recording device, further in particular by means of tracking, a cost function in one embodiment can be changed to c = sum(amount(traj-sim - traj_real)), where traj_sim refers to a recorded trajectory of the object in the simulation environment, in particular describes this trajectory, and where traj_real refers to a recorded trajectory of the object in reality, in particular describes this trajectory, or this cost function can be applied for optimization. In one embodiment, in particular by recording the object with the recording device, at least one start position of the object, in particular a start frame, and one end position of the object, in particular an end frame, can be used for a cost function.
[0032] In one embodiment, this advantageously makes it possible for the parameter estimation for the optimization to be or will be improved, in particular because the trajectory of the object contains more information than, in particular, a gripping success variable that indicates or describes the success or failure of the grip.
[0033] In one embodiment, alternatively or additionally, a movement of the object can be approximated via the end position of the holding robot and in particular can be included in the cost function.
[0034] In one embodiment, this advantageously allows information about the gripping process to be obtained that goes beyond the information content of the gripping success, in particular can or does offer an improved cost function.
[0035] In one embodiment of the invention, a method for controlling a gripping robot is provided. In one embodiment, the method comprises determining control data based on the optimized parameters according to an embodiment described herein, in particular for robot-assisted gripping, further in particular for a robot-assisted gripping process. In one embodiment, the method comprises controlling and / or moving the gripping robot based on the optimized control data.
[0036] This makes it possible, in one embodiment, for a movement and / or control of the robot to be first optimized in a simulation environment and then transferred to the robot, in particular in order to grip an object to be gripped better and / or more quickly. Furthermore, in one embodiment, this makes it possible for the grip, in particular a closing force of the gripper fingers or the like, to be used in a more optimized manner. In one embodiment, embodiments of a method described herein are applicable or transferable to applications with multiple objects, in particular applications with multiple objects to be gripped in a container, provided that this is technically reasonable and / or applicable.In particular, in one embodiment, the parameters determined for exposed objects can be transferred, in particular applied, to a gripping process with several objects in a container, and further, in particular based on the optimized parameters, control data for a gripping robot can be determined which is intended to grip or grips at least one object from several objects, in particular arranged in a container.
[0037] In one embodiment of the invention, a system for operating and / or monitoring at least one robot is provided. In one embodiment, the system comprises a gripping robot and means for determining a plurality of, in particular collision-free and / or graspable, grip positions, in particular a grip sampler. In one embodiment, the system and / or its means comprise means for selecting one of the determined grip positions, in particular a processing unit configured to select one of the determined grip positions. In one embodiment, the system and / or its means comprise means for gripping the object. In one embodiment, the system and / or its means comprise means for storing the gripping success, in particular a processing unit and / or in particular a memory, in particular in data connection with the processing unit.
[0038] In one embodiment, the system has means for determining real gripping data and / or means for determining real holding data, in particular at least one sensor that is configured to detect at least one parameter relevant to the real gripping data and / or to detect a parameter relevant to the real holding data. In one embodiment, the system has means for simulating, in particular for replicating, at least one gripping process, in particular grip, of a gripping robot in a simulation environment. In one embodiment, the system has means for data-based optimization of parameters, in particular of the simulation environment, in particular a processing unit. In one embodiment, the system has a gripping robot and a holding robot. In one embodiment, the system has means for determining a holding grip position, in particular a processing unit.In one embodiment, the system, in particular the holding robot, comprises means for gripping, in particular holding, the holding grip position. In one embodiment, the system, in particular the holding robot, comprises means for exerting a force on the object.
[0039] In one embodiment, the system has a recording device. A recording device, as described herein, particularly preferably comprises a recording device for recording digital and / or two-dimensional, in particular three-dimensional, images, and can in particular have at least one 2D camera, 3D camera and / or at least two spatially spaced-apart cameras and / or at least one scanner, preferably for three-dimensional scanning. In one embodiment, the recording device is configured to record a point cloud and / or color information, preferably a three-dimensional point cloud, more particularly a point cloud with color information, in particular assigned to the points of the point cloud. Accordingly, in particular a three-dimensional point cloud and / or color information recorded with the aid of a recording device is stored as a frame orImage, which in one embodiment is generally a three-dimensional image and / or an image with color information.
[0040] In one embodiment, the receiving device is arranged on at least one of the robots. In one embodiment, the receiving device is alternatively or additionally arranged remotely from the robot, in particular such that the receiving device can receive, in particular track, an object to be grasped and / or a grasped object.
[0041] A system and / or means within the meaning of the present invention can be designed in hardware and / or software, in particular at least one, in particular digital, processing unit, in particular a microprocessor unit (CPU), graphics card (GPU) or the like, preferably connected to a memory and / or bus system for data or signals, and / or one or more programs or program modules. The processing unit can be designed to execute instructions implemented as a program stored in a memory system, to detect input signals from a data bus, and / or to output output signals to a data bus. A memory system can have one or more, in particular different, storage media, in particular optical, magnetic, solid-state, and / or other non-volatile media. The program can be designed in such a way that it embodies the methods described here orcapable of carrying out such a method, so that the processing unit can carry out the steps of such methods and thus in particular operate or monitor the robot.
[0042] In one embodiment, a computer program product can comprise, in particular be, a storage medium, in particular a computer-readable and / or non-volatile one, for storing a program or instructions or with a program or instructions stored thereon. In one embodiment, execution of this program or these instructions by a system or a controller, in particular a computer or an arrangement of multiple computers, causes the system or the controller, in particular the computer(s), to carry out a method described here or one or more of its steps, or the program or the instructions are configured to do so.
[0043] In one embodiment, one or more, in particular all, steps of the method are carried out completely or partially automatically, in particular by the controller or its means.
[0044] Further advantages and features emerge from the subclaims and the exemplary embodiments. The following shows, partly schematically:
[0045] Fig. 1 : A system according to an embodiment of the present invention;
[0046] Fig. 2: a system according to an alternative or additional embodiment; and Fig. 3: a schematic block diagram of a method according to an embodiment.
[0047] Fig. 1 schematically shows system 1 with a gripping robot 2 having a gripper 3 for gripping an object 5 to be gripped. The object 5 to be gripped is shown on a schematic worktop 6 of the system 1, with a single object 5 to be gripped being shown in solid lines. In one embodiment, the system 1 is configured to grip multiple objects 5, 5' (shown in dashed lines) in a container (not shown), in particular to carry out a method described herein. In one embodiment, the system 1 can have a receiving device 8 which, as shown in dashed lines in Figure 1, is not arranged on the robot 2. In one embodiment, the receiving device can be arranged on the robot 2, in particular on a flange of the robot 2.Furthermore, the system in Figure 1 has a processing unit 7, which is in data communication with the robot 2 and is configured in particular to control and monitor the robot 2. The robot 2 is further configured to grip the object 5 to be grasped based on a determined grip position. If the gripping process of the robot 2 is repeated and the gripping success of the grip position is recorded, the gripping sequences can be ordered according to the grip position and / or the pose, in particular location, of the object and, in particular based on this recording, a sorting of the grip positions according to gripping success can be derived. For this purpose, in particular the grip position that has (so far) been grasped the least can be selected. Accordingly, data, in particular on the gripping success of possible grip positions, which are or were determined in particular using a grip sampler, can be determined and stored.These data can be used, in particular, for a simulation of the system 1 in a simulation environment, wherein the system 1 in the simulation environment comprises, at least substantially, the components of the real system 1. Thus, the data obtained from the real system 1 can be used to optimize the parameters in the simulation environment, in particular based on the data determined in reality, in particular real gripping data.
[0048] Fig. 2 schematically shows a system 1 with a holding robot 10. Furthermore, Figure 2 shows a robot 2 which has gripped an object 5 to be grasped. Figure 2 shows, indicated in dashed lines, comparable to Figure 1, a receiving device 8 and also a processing unit 7 which is data-connected to the robot 2 and in particular to the holding robot 10 (also shown in dashed lines). The holding robot 10 is shown in Figure 2 in such a way that it has gripped a determined holding grip position and is holding the object already gripped by the gripping robot in the direction opposite to the gripping direction of the gripping robot 2. In some embodiments, the holding robot 10 can compensate for gravity, particularly in the case of heavy and / or large objects.In Figure 2, it is further indicated by arrows that the holding robot 10 can apply a force to the object 5, in particular successively, which acts counter to a force exerted by the robot 2. In some embodiments, such a force can be rotational (opposite the rotational force of the robot 2) and, in particular, can be successively increased until the robot 2 loses the object or drops it. In some embodiments, values for parameterizing the simulation environment or the simulation of the grip can be derived from the maximum force used, and in particular these values can be transferred to the simulation. This is indicated in Figure 2 by the data connections in dashed lines to the processing unit 7. Furthermore, the gripped object 5 can be moved during the process orMethod can be tracked by a recording device 8, in particular a trajectory of the object can be recorded, in embodiments in particular continuously and / or in time steps. The recording device 8 can, as indicated in Figure 2 by the dashed line, be attached externally or not be arranged on at least one of the robots 2, 10, or in embodiments can be arranged on a flange of at least one robot 2, 10. The system 1 shown in Figure 2 can then be constructed, at least substantially, identically in the simulation environment, so that a simulated (held) grip on an object can be simulated or is simulated to the grip on the object in reality, in particular in embodiments is repeated until the simulated grip resembles the real grip, at least substantially (or within predetermined limits for an accuracy of reproduction).
[0049] Figure 3 schematically shows a block diagram of a method 30. The method comprises a step S1 for determining data (in reality). In this case, real gripping data S10 can be determined and / or real holding data S20 can be determined. The determination of real gripping data S10 comprises, in particular, a determination of grip positions on the object 5 to be grasped, represented by S12, wherein, in embodiments, a randomized placement of the object to be grasped can be or is preceded by this. Furthermore, a step S14 for selecting one of the determined grip positions is schematically represented in Figure 3. S16 in Figure 3 relates, in particular, to gripping the object 5 based on the grip position selected in S14. S18 schematically represents a storage of the gripping success for the selected grip position, in particular in a database or memory. These steps can be repeated, as represented in particular by the dashed arrow.In some embodiments, real gripping data can also be determined independently and optimization can be carried out, particularly based solely on the stored data on gripping success, to improve or optimize the grip sampler.
[0050] Furthermore, in Figure 3, S20 schematically illustrates the determination of real holding data. In the embodiment shown, the determination of real holding data S20 comprises determining a holding grip position S22 on an object 5 gripped by a gripping robot. As previously indicated in Figure 2, the grip position of the holding robot 10 is arranged such that, in one embodiment, a force can be applied to the object that is opposite to a force on the object 5 that is or is applied by the gripping robot 2. Furthermore, as shown by way of example, a gripping S24 follows at the determined holding grip position, in particular by the holding robot 10. This is followed, as shown by way of example in Figure 3, by exerting a force S26, in particular on the object 5, in the opposite direction to a force of the gripping robot 2 acting on the object 5.
[0051] In embodiments, the determination of holding real data S20 can follow the gripping of the object S16 or the storage S18 of the determination of gripping real data S10, as indicated in particular by the dashed arrow between S18 and S20.
[0052] Furthermore, Figure 3 shows a simulation S30 of at least one of the preceding or above-described (real) gripping processes. During simulation S30, one embodiment attempts to replicate the real gripping process, in particular as closely as possible, within a simulation environment. S32 in Figure 3 illustrates an optimization of the parameters by way of example. The optimization S32 is based on the previously determined real gripping data and / or the determined real holding data. Furthermore, as shown in dashed lines, the method can comprise a step of controlling, moving, and / or monitoring S40 a robot 2, 10, in particular during gripping, based on the determined optimized parameters, which for this purpose are transmitted in one embodiment as control data for the robot 2, 10. Although exemplary embodiments were explained in the preceding description, it should be noted that a multitude of modifications are possible.Furthermore, it should be noted that the exemplary embodiments are merely examples and are not intended to limit the scope of protection, applications, or structure in any way. Rather, the preceding description provides the skilled person with a guide for implementing at least one exemplary embodiment. Various modifications, particularly with regard to the function and arrangement of the described components, may be made without departing from the scope of protection as defined by the claims and equivalent combinations of features.
[0053] List of reference symbols
[0054] 1 system
[0055] 2 gripper robots
[0056] 3 grippers
[0057] 5, 5' object to be grasped
[0058] 6 Work surface
[0059] 7 Processing unit
[0060] 8 Mounting device
[0061] 10 holding robots
[0062] 30 procedures
[0063] S10 Determination of real gripping data
[0064] S12 Determining grip positions
[0065] S14 Selecting a determined grip position
[0066] S16 Grasping the object
[0067] S18 Saving the gripping success
[0068] S20 Determination of real holding data
[0069] S22 Determining a holding grip position
[0070] S24 Gripping at the holding grip position
[0071] S26 Exerting a force
[0072] S30 Simulate
[0073] S32 Optimize
[0074] S40 Controlling / moving a robot
Claims
Patent claims Method (30) for automatically optimizing parameters for a robot-assisted gripping process, comprising - Determining real gripping data (S10), wherein the real gripping data (S10) describe at least one gripping success of a gripping position on an object gripped by a gripping robot; and / or - Determining real holding data (S20), wherein the real holding data (S20) describe at least one parameter associated with a gripping process, in particular gripping, in particular a force on the object gripped by a gripping robot; - Simulating (S30), in particular replicating, at least one gripping process, in particular grip, in a simulation environment based on the gripping process, in particular grip, underlying the determined gripping real data (S10) and / or the determined holding real data (S20); - Data-based optimization (S32) of parameters, in particular optimization of the parameters based on the determined real gripping data (S10) and / or based on the determined real holding data (S20), to determine optimized parameters. Method (30) according to the preceding claim, characterized in that the data-based optimization (S32) comprises an automatic adjustment of the parameters, in particular a simulated mass, simulated dynamic parameters, and / or a simulated force, in particular friction.Method (30) according to one of the preceding claims, characterized in that the data-based optimization (S32), in particular the optimization of the parameters, is based on a cost function of real gripping data (S10) and gripping simulation data and / or real holding data (S20) and holding simulation data, wherein the gripping simulation data describe at least one simulated gripping parameter, in particular a gripping success of a gripping position in the simulation, and wherein the holding simulation data describe at least one simulated holding parameter, in particular a force of the simulated grip.
4. Method (30) according to one of the preceding claims, characterized in that the optimization is carried out by means of a gradient-free optimizer.
5. Method (30) according to one of the preceding claims, characterized in that the determination of gripping real data (S10) comprises: - Randomized placement of an object to be grasped; - determining (S12) several, in particular collision-free and / or graspable, grip positions on the object to be grasped; - Selecting (S14) one of the determined grip positions; - gripping (S16) the object, in particular with a gripping robot, based on the selected grip position; - Saving (S18) the gripping success to the selected grip position - Repeating the steps until a predetermined termination criterion and / or until a predetermined number of repetitions, in particular in order to obtain reality data, in particular a pre-sorted set of grips, further in particular according to gripping success.
6. Method (30) according to the preceding claim, characterized in that the selection (S14) is based on a selection frequency of the determined grip positions, and in particular the determined grip position with the lowest selection frequency is selected.
7. Method (30) according to one of the preceding claims 5 or 6, characterized in that the randomized placement is carried out by a gripping robot, in particular that the object is moved by the gripping robot to a random position and is then dropped.
8. Method (30) according to one of the preceding claims, characterized in that the determination of holding real data (S20) comprises: - Determining a holding grip position (S22), wherein the holding grip position is arranged on the object gripped by the gripping robot in such a way that a force can be exerted on the gripped object via the holding grip position, which acts opposite to a force on the grasped object, in particular opposite to a gripping direction of the gripping robot. - gripping (S24), in particular holding, the object at the determined holding grip position; - Exerting a force (S26) on the object, which acts in particular opposite to the gripping direction of the gripping robot. Method (30) according to the preceding claim, characterized in that the exertion of the force comprises successively increasing the force in predetermined steps, in particular until the gripping robot loses the object and / or a maximum force is exceeded. Method (30) according to one of the preceding claims, characterized in that determining real gripping data (S10) and / or real holding data (S20) comprises picking up the object by means of a picking device, in particular determining a pose of the object, in particular over time. Method (30) for controlling a robot to perform a robot-assisted gripping process, comprising: - Determining control data based on the optimized parameters according to one of the preceding claims 1 to 10, - Controlling and / or moving (S40) the robot based on the optimized control data to perform a robot-assisted gripping process. Method (30) according to one of the preceding claims, characterized in that the determined optimized parameters are transferred, in particular applied, to a gripping application with multiple objects in a container. System (1) for operating and / or monitoring at least one robot, in particular a gripping robot, which is configured to perform a method (30) according to one of the preceding claims. Computer program or computer program product, wherein the computer program or computer program product contains instructions, in particular stored on a computer-readable and / or non-volatile storage medium, which, when executed by one or more computers or a system (1) according to claim 13, cause the computer(s) or the system (1) to carry out a method (30) according to one of claims 1 to 12.