Predefining the maximum permitted speed of a robotic device

Abstract

The invention relates to predefining a maximum allowable speed of a robotic device (1), comprising: predefining a contact point between a human operator and the robotic device (1), a shape of the robotic device (1) at the contact point and spatial boundary conditions of the collision for a collision between the human operator and the robotic device (1) in order to predefine a maximum allowable speed of the robotic device (1) in the most efficient way possible such that a maximum possible maximum allowable speed can be determined, at a speed which ensures that biomechanical limits for preventing injuries are respected; determining, taking into account the spatial boundary conditions, using a computing unit (4) whether the collision is a non-clamping collision or a clamping collision; calculating by the computing unit (4) a maximum allowable speed of the robotic device (1) at the contact point using a free impact model if the collision is a non-clamping collision and using a clamping impact model or using a quasi-static clamping model if the collision is a clamping collision, wherein the models are different in each case; and outputting, using the computing unit (4), a signal which is dependent on the calculated maximum allowable speed of the robotic device (1).

Description

[Technical field]

[0001] The present invention relates to a computer-implemented method for predefining a maximum allowable speed of a robotic device, and a corresponding control unit. [Background technology]

[0002] In order to prevent or reduce accidents and injuries with robotic devices, it is necessary to protect mechanical human-machine interfaces, including moving parts. Such human-machine interfaces exist, for example, in collaborative robotic devices / robots, so-called cobots. In principle, there exists a risk of injury resulting from a collision between each moving part of the robotic device (machine), as referred to herein, and parts of the body of the human operator (user). The risk of injury or accident can be reduced by limiting the performance of the robotic device to such an extent that the device does not pose a risk of injury or only poses a reduced risk of injury.

[0003] In general, performance limitations may be achieved by limiting the force and / or kinetics of the robotic device. For example, known biomechanical limits quantified in relevant standards accordingly indirectly specify, with respect to a safe operating mode of the robotic device, to what extent the kinetics of the robotic device, e.g. the speed of the moving parts of the robotic device, should be limited so that a collision with a human does not cause an intolerable overstress of the human tissue and consequent pain or injury.

[0004] Nowadays, the verification of limit values ​​in robotic devices by physical measurements, i.e. compliance with limit values, is established. The drawback of this verification is that the corresponding measurements are only possible after the robotic device has been put into operation, and therefore the maximum achievable speed or performance limits can only be determined retrospectively. As a result, the corresponding permissible control signals can also only be generated retrospectively, which leads to long iteration loops and a high degree of planning uncertainty.

[0005] The theoretical understanding of human-resilient robotic device collisions abstract assessment was significantly influenced by the article "Fast and "Soft Arm" Tactics" by Bicchi, A. et al., published in IEEE Robot.Automat.Mag.11(2), pp. 22-23, 2004, where a three-mass oscillator model was used to propose the risk of head trauma. However, this model is only suitable for rigid body parts such as the head.

[0006] The three-mass oscillator model is similarly used in the article "On Impact the Decoupling Properties of Elastic Robots and Time Optimal Velocity Maximization on Joint Level" by Haddadin S. et al., published in Proceedings of 2012 IEEE / RSJ-International Conference on Intelligent Robots and Systems, pages 5089-5096. To illustrate the worst-case scenario, the drive inertia is assumed to be of infinite mass, which would lead to significantly excessive collision forces and therefore unnecessarily reduced maximum allowable velocities in realistic scenarios.

[0007] The article "A New Approach to Estimate the Apparent Mass of Collaborative Robot Manipulators" by Herbster S. et al. in "Experimental Robotics", 2021, Cham: Springer International Publishing, pages 211-221, also uses the three-mass oscillator model. There, the analytical solution for the safe speed, i.e. the maximum allowable speed, is described, assuming a linear characteristic curve. Summary of the Invention [Problem to be solved by the invention]

[0008] The objective is therefore to specify a maximum permissible speed for a robotic device, ensuring that biomechanical limits to prevent injuries are respected in the most efficient way possible, so that the highest possible maximum permissible speed can be determined, especially in real-time applications.

[0009] This object is achieved by the subject matter of the independent claims. Advantageous embodiments can be found in the dependent claims, the description and the figures. [Means for solving the problem]

[0010] One embodiment relates to a computer-implemented method for predefining an allowable maximum speed of / for a robotic device. In this specification, the maximum speed is particularly related to one or more points moving with the highest speed among all points on the surface of the robotic device at each time with a predefined machine path or trajectory of the robotic device. In particular, the robotic device may be a collaborative robotic device, a so-called cobot. In general, however, the described method may be applied to any machine having a physical interface with humans.

[0011] Herein, a method step is to predefine at least one contact point between a human operator and a robotic device, a shape of the robotic device at the contact point, and a spatial boundary condition of the collision for a collision between the human operator and the robotic device. In particular, the shape of the robotic device may include data of a tool and / or a workpiece. Thus, a tool used by the robotic device or a workpiece processed by the robotic device may be considered as part of the robotic device. The selection of the contact point between the human and the machine, i.e. between the operator and the robotic device, and the spatial boundary conditions, e.g. whether the part of the human body belonging to the contact point is captured or free, may be performed manually by a user or (partially) automatically. A user or (partially) automated input may specify one or more contact points of a collision or a multi-contact scenario, each including an associated collision of the operator and the robotic device. The shape of the robotic device associated with the collision may be selected manually by a user, e.g. using a list, or automatically, e.g. based on a 3D model of the robotic device, possibly including the tool and the workpiece.

[0012] In a further method step, using the computing unit, it is determined whether the collision is a non-clamping or a clamping collision, taking into account predefined spatial boundary conditions. In case of a collision with multiple contact points, a collision may occur that is both a non-clamping and a clamping collision. In this case, the collision may be split, for example, into two separate partial collisions for the method, and then for each partial collision the described method is carried out separately. At the end of each separate method, as with the predefinition of the maximum permissible speed of the robotic device for multiple collisions or multiple contact points, as described further below, the different results of the maximum permissible speed may be compared with each other, and a suitable, for example lowest, maximum speed may be predefined as the maximum permissible speed, i.e. calculated and provided for output, as described below.

[0013] Thus, a further method step is the calculation of the maximum permissible speed of the robotic device at the contact point. This calculation is performed using a free-impact model if the collision is a clamping collision, using a clamping-impact model if the collision is a clamping collision and / or using a quasi-static-clamping model. These models are different in each case. The free-impact model is used here for "free impacts", i.e. relatively high-speed collisions in which the contact point at the part of the human body or the part of the human body belonging to the contact point is not hindered by any external resistance in the impact direction (direction of movement of the contact point on the robotic device during the collision), i.e. the part of the human body can avoid it. The clamping-impact model is related to similarly relatively high-speed collisions in which the part of the human body assigned to the contact point cannot avoid it, i.e. it encounters an external resistance and is therefore captured. A quasi-static clamping model is used in this case for low-speed collisions compared to the high-speed collisions discussed above, in which the contact point on the body part or the body part associated with the contact point still experiences an external resistance in the direction of the collision, i.e. the body part is unable to avoid the clamping impact.

[0014] Herein, different models may be stored in a computing unit that calculates the maximum permitted speed, the computing unit determining each maximum speed specific to the model based on partly identical and partly different input variables, as further explained below. Herein, whether the impact is relatively fast or relatively slow, i.e. whether the clamping impact model or the quasi-static clamping model is used, may be predefined, for example, by setting corresponding speed limit values, alternatively or additionally, by applying both models, i.e. by simulating the impact using two different simulations separately and then comparing the results and obtaining the maximum permitted speed in each case. For example, the lower of these results may be selected for further parts of the method.

[0015] Finally, the computing unit outputs a control signal, which may be selected from a plurality of calculated maximum velocities, depending on the calculated maximum allowed velocity of the robotic device, in particular representing the calculated maximum allowed velocity. This control signal may here be a control signal that can be read directly by the robotic device, or alternatively may be a control signal that accordingly indicates to a human user of the method the maximum velocity at which the robotic device should be controlled, for example via a display unit. Thus, the computer-implemented method may be implemented as a program sequence in a virtual planning tool, for example using a virtual robotic device that is virtually controlled by the control signal, or directly in the control unit of the robotic device. In addition to the tool, the safety configuration, the description of the kinematics, the mass, the inertia, the center of gravity, the joints and the corresponding motors, on the other hand, the required data regarding the torques of the robotic device, as well as the limit values, stiffness characteristic curves and masses of the body parts affected by the collision, are stored in a corresponding database and provided to the method. In the following, the data used are explained in more detail. The signal may for example be a warning signal, indicating that in a real-time application, the actual velocity of the robotic device is higher than the maximum allowed velocity.

[0016] To calculate the maximum allowable speed, the method accesses parameters stored in a database or model. After the calculation is completed, the method and therefore the algorithm outputs a safe speed, i.e., a maximum allowable speed of the robotic device, at which the considered mechanical parts involved in the collision, in case of the considered collision, do not exceed the stored biomechanical limit values ​​that are acceptable according to a standard valid in July 2021, such as ISO / TS 15066. In case of multiple pairs of contact points and contact shapes in one or more collisions, the lowest value of the calculated safe speeds may be selected for output.

[0017] The advantage here compared to previous approaches is that firstly, clamping impact and quasi-static clamping are combined in a uniform manner, but still considered separately, allowing to calculate exactly which speeds are still safe with low risk of injury, both for relatively slow and relatively fast impacts.

[0018] This is based on the knowledge that the behavior of the system is highly nonlinear, i.e. that the constraints on a uniform model for all tightenings, fast and slow tightenings, tightening impulses and quasi-static tightenings, result in permissible maximum speeds that deviate significantly from the real limit values ​​for the other cases and are therefore very inaccurate. In addition, the described method is suitable for both so-called offline (planning) and real-time calculations of the respective permissible maximum speeds of machines, in particular robotic devices with a physical interface with humans. Herein, a particular example is constituted by a collaborative robotic device that communicates with a human and operates in a power and force limiting mode according to ISO / TS15066. The described method may be used to ensure compliance with predefined limit values, such as known biomechanical limit values, thus minimizing the risk of unacceptable stresses, i.e. collisions causing pain and / or injuries to the operator.

[0019] The aforementioned method may be used to optimize and verify the permissible speeds of the moving parts of a machine along a predefined trajectory, or more precisely, a path (trajectory) of the machine traversed by a robotic device. Different spatial and temporal conditions of collisions can be specifically addressed, which results in the outstanding accuracy of the method.

[0020] A further insight underlying the present invention is therefore that the use of various models is more beneficial than the use of a single model in terms of overall efficiency, although the required initial configuration is more complex. The method can also take into account the various surface geometries at the contact point of the colliding machine parts, including the tool and the workpiece, as well as the non-linear stiffness characteristic curves at the contact point for both the loaded soft tissue of the operator, which may also be deformed in the event of a collision, and the machine surface at the contact point. The method may be applied at various points along the machine surface, including the tool and the workpiece, in particular so that a comprehensive assessment of the risk of injury and therefore, in particular, a reliable determination of the maximum permissible speed is achieved.

[0021] In an advantageous embodiment, it is provided that the output control signal represents a position-dependent speed specification along a predefined mechanical path for the robotic device, a trajectory of the robotic device, or one or more portions of the robotic device. The speed specification is herein position-dependent in that it is predefined along the mechanical path and thus at various positions of each mechanical portion of the robotic device. In particular, the position-dependent speed specification may be generated by modifying the magnitude of an originally predefined speed specification for the predefined mechanical path, which is preferably also position-dependent based on a calculated maximum speed. Thus, an original position-dependent speed specification along the mechanical path that may be unsafe may be replaced with an equivalent position-dependent speed specification that ensures that a safe allowable maximum speed is predefined along the predefined mechanical path.

[0022] In this specification, the change in magnitude may include or be a uniform, i.e. position-independent, change in magnitude of the predefined maximum speed, or a change in magnitude that is locally adapted to the predefined machine path. Thus, in a uniform change in magnitude, the predefined maximum permitted speed is multiplied by a fixed factor depending on the position along the machine path, so that at one or more speed extremes of the original speed specification, a safe speed at which injuries can be excluded as described above is not exceeded. In the case of a change in magnitude that is locally adapted to the predefined machine path, the speed may be reduced accordingly in the case of subsections of the machine path where the original speed specification exceeds the safe speed, and the speed may be increased in the case of subsections of the machine path where the originally predefined speed specification falls below the determined safe maximum speed. In particular, the predefined process speed of the robotic device may be taken into account in one or more subsections of the machine path, for example, in such a way that the originally predefined maximum speed specification is not adjusted there. This is advantageous, for example, if in such a subsection of the technological process carried out by the robotic device a lower speed is required with respect to the calculated maximum permitted speed for the process itself, which may be increased for safety reasons. This has the advantage that the entire process of the robotic device can be safely optimized, i.e. the speed of the robotic device can be increased.

[0023] In an alternative embodiment, it is provided that the maximum allowable speed is calculated in real time and the control signal output represents the instantaneous maximum allowable speed of the robotic device, such that the instantaneous actual speed of the robotic device at the predefined contact point is adapted to the instantaneous maximum allowable speed. This is advantageous, for example, in the case of manually or semi-automatically controlled devices, where the instantaneous speed of the robotic device is controlled or influenced by an operator.

[0024] In a further advantageous embodiment, it is provided that if the collision is a clamping collision, the maximum permitted speed is calculated using the clamping impact model and the quasi-static clamping model, and the lower of the calculated maximum speeds is selected as the maximum permitted speed on which the output signal depends. This has the advantage that an implicit distinction is made between faster and slower collisions, i.e. inflexible and possibly incorrect limit values ​​do not have to be set from the start, but safer maximum speeds are given priority.

[0025] In a further advantageous embodiment, the maximum permitted speed is calculated with a quasi-static tightening model based on: (a) A predefined kinematic structure of the robotic device, (b) the joint configuration, including, inter alia, the positions of one or more axes of the robotic device and the velocities assigned to each axis, at the time of the collision; (c) stiffness characteristic curve of the human body part at the contact point of the impact; (d) stiffness characteristic curve of the machine point at the impact contact point; (e) a resultant stiffness characteristic curve calculated using the stiffness characteristic curve of the body part and the stiffness characteristic curve of the machine point; (f) a maximum deformation, predefined and added using or from, in particular, a biomechanical force threshold and / or an energy threshold and / or a deformation threshold; (g) the stiffness characteristic curve of the machine points and the allowable penetration depth calculated using the resulting stiffness curve; (h) reaction force of the robotic device, (i) The reaction distance of the robotic device calculated using the stiffness characteristic curve of the machine point and the allowable penetration depth, (j) the allowable braking distance of the robotic device calculated using a predefined maximum allowable deformation and reaction force, and (k) The actual braking distance of the robotic device.

[0026] The predefined kinematic structure of the robotic device according to (a) represents herein the global arrangement of the joints, for example as Denavit-Hartenberg parameters. The joint configuration at the time of the collision according to (b) is in particular triggered by, for example, a human input, by a corresponding control unit of the robotic device or a virtual (offline) planning tool, which transmits the axial positions and axial velocities of the robotic device to the method or algorithm. If the trajectory of the robotic device is transmitted, then the joint configuration is calculated and transmitted accordingly for each individual time step. Thus, the method may then be used to calculate a safe speed for each time step along the trajectory.

[0027] The stiffness characteristic curve of the body part according to (c) depends on the geometry of the machine at the affected body part and the contact point. The stiffness characteristic curve quantifies the dependency of the forces acting on the body part during the collision on the deformation of the body part in question and is usually distinguished by a non-linear behavior. The stiffness characteristic curve of the body part affected by the collision and the associated limit values ​​may be selected from a tabular list, determined based on a contact model (simulation) or determined experimentally, in particular according to ISO / TS15066 here. Known tabular lists include, for example, the affected body part and the outline of the geometry of the robotic device at the contact point. On the other hand, the contact model may calculate the stiffness characteristic curve and the associated force or energy limit values ​​depending on the contact geometry based on specific material parameters and biomechanical limit values ​​per body part or contact point at the body part.

[0028] The machine point at the contact point according to (d) may have a stiffness comparable to that of the body part. Similar to the stiffness characteristic curve of the body part, the corresponding stiffness characteristic curve may be given by an entry in a database, a simulation, or experimental data. This stiffness characteristic curve may represent a non-linear behavior. The effective stiffness characteristic curve according to (e) is obtained from the stiffness characteristic curve of the body part according to (c) and the stiffness characteristic curve of the machine point at the contact point according to (d). Such a resulting stiffness characteristic curve is shown in FIG. 2, for example, and typically establishes a relationship between the force acting on the body part at the contact point in the event of a collision with the machine point and the deformation of the body part at the contact point resulting from this force. Herein, if the acting force exceeds the acceptable biomechanical contact force, the acceptable penetration depth is also exceeded, resulting in damage, i.e. injury to the body part.

[0029] The predefined maximum allowable deformation according to (f) as a biomechanical limit is a limit that depends on the body part and the contact geometry. Thus, the maximum allowable deformation may exist together with or as a force threshold / limit and / or an energy threshold / limit and / or a deformation threshold / limit. In this specification, a distinction may be made between fast (transient) and slow (quasi-static) values ​​for each limit of the impact load and the clamping load. The quasi-static designation or value relates to a quasi-static clamping model, while the temporary limit relates to a free impact model or a clamping impact model. Using the specific stiffness characteristic curve of the human body part, the various force, energy or deformation limit values ​​may be converted into each other. For example, as explained in conjunction with FIG. 2, the biomechanical force limit value is required for this calculation.

[0030] The allowable penetration depth according to (g) may be determined from the biomechanical force limit and the resulting stiffness characteristic curve. In FIG. 2, the allowable penetration depth corresponds to the intersection of the allowable biomechanical contact force and the resulting stiffness characteristic curve. The reaction force of the robotic device according to (h) is a force limit that may be set for the robotic device, which, if exceeded, will cause the robotic device to trigger a safety stop. Such a safety stop typically involves asynchronous braking using the maximum deceleration of all axes, in which case path fidelity is not provided. Exemplary reaction forces are also shown in FIG. 2. The configurable force limit may often be set for the operating point of the robotic device across a particular axis. The force limit may be set as an axis-specific force or torque limit, in particular, shown for each drive of the robotic device. The axis-specific force or torque limit may be converted to a global force limit using the axis position, the point of the machine at the contact point, i.e., the contact point on the surface of the machine, and the collision direction.

[0031] In principle, force limits may be defined for any machine point, i.e. for any contact point on the machine surface. Force limits set at a machine point, e.g. an end effector of a robotic device, may usually be transformed into force limits for any other machine point. Then, axis-specific force or torque limits do not have to be given for all axes. If the axis position is unfavourable, the robotic device may not detect a collision in such a case. This is the case if none of the monitored axis-specific forces or torques are addressed by the input forces in case of a collision. The operating point force limits and the axis-specific force or torque limits, both commonly called monitored limits, are often both set in the robotic device, i.e. predefined and activated. In such cases, the lower of the two resulting limits has to be taken into account as a reaction force in the robotic device for further consideration.

[0032] The calculated reaction distance of the robotic device according to (i) may be obtained from the reaction forces set at the operating point of the machine and the resulting stiffness characteristic curve. The intersection of the limit values ​​and the characteristic curve then corresponds to the length of the reaction distance, as also shown in FIG. 2. The allowable braking distance of the robotic device according to (j) is obtained from the allowable penetration depth minus the reaction distance of the machine. An exemplary allowable braking distance is also shown in FIG. 2. If the reaction forces on the robotic device are greater than the allowable contact forces at each body position, there is no safe speed. In this case, the robotic device must not move until the affected body parts are outside the danger zone.

[0033] In case of a safety stop, the robotic device usually performs an asynchronous braking, in which all axes are decelerated to the maximum and a loss of path accuracy is accepted. Herein, the position of the axes of the robotic device after such a braking process is calculated from the speed of the axes at the time of the collision, the position of the axes whose reach can be determined, the additional load at the point of application or tool, and the braking angle traveled. The braking angle may be chosen from the manufacturer's specifications or calculated based on the possible decelerations or determined experimentally. The deceleration is then obtained from the forces or torques available at the axes of the robotic device and the dynamics of the robotic device considered as a mechanical system. Using the known kinematic structure and taking into account the difference in the distance of the collision point before and after the end of the braking process, the distance traveled in the collision direction may be calculated for any point of the machine, i.e. for each point of the machine as the contact point of the collision. This traveled distance then corresponds to the actual braking distance of the machine according to (k).

[0034] In an advantageous embodiment, the maximum permitted speed may be determined iteratively based on the reaction forces of the robotic device until the actual braking distance corresponds to the permitted braking distance. Here, the bisection method is particularly suitable. The maximum permitted speed may be calculated particularly quickly using the aforementioned variables due to fewer iteration steps.

[0035] In another advantageous embodiment, if the collision involves contact points at different parts of the human body, the control signal output depends on the maximum allowable speed, which corresponds to the shortest actual braking distance, based on a predefined maximum allowable deformation and the resulting stiffness curve, and which does not necessarily have to be the lowest maximum allowable speed at the various contact points. This has the advantage that injuries are avoided particularly effectively, since it is not the maximum speed itself that is taken into account, but each contact point of the collision that is evaluated with respect to the required braking distance.

[0036] In a further advantageous embodiment, it is provided that if the collision involves contact points at different machine points, the permissible maximum speeds of all contact points are calculated and the output signal depends on the lowest permissible maximum speed. In contrast to what was explained in the previous paragraph, it has been found that for different machine points, the maximum speed itself is best suited to limit the risk of injury, and not the actual braking distance.

[0037] In a further advantageous embodiment, it is provided that in the case of the free impact model and the clamping impact model, the permissible maximum velocity is calculated in each case on the basis of: (a) A predefined kinematic structure of the robotic device, (b) the joint configuration, including, inter alia, the positions of one or more axes of the robotic device and the velocities assigned to each axis, at the time of the collision; (c) stiffness characteristic curve of the human body part at the contact point of the impact; (d) stiffness characteristic curve of the machine point at the impact contact point; (e) a resultant stiffness curve calculated using the stiffness characteristic curve of the body part and the stiffness curve of the machine point; and (f) Maximum allowable deformation predefined using biomechanical force threshold and / or biomechanical energy threshold.

[0038] The values ​​stated correspond to those already explained above with respect to the quasi-static tightening model.

[0039] In addition, an additional maximum speed is calculated herein based on: (l) the effective mass of the machine at the point of contact of the impact, and (m) Effective stiffness of the robotic device.

[0040] Using only the free impact model, the maximum velocity is calculated based also on: (n) The effective mass of a body part at the point of contact in a collision.

[0041] The effective mass of a machine point according to (l) is calculated here according to Khatib O. and can be found accordingly in the article "A Unified Approach for Motion and Force Control of Robotic Manipulators: The Operational Space Formulation" published in 1987 in IEEE Journal of Robotics and Automation 3(1), pages 43-53 and in the article "Inertial Properties in Robotic Manipulation: An Object-Level Framework" published in 1995 in The International Journal of Robotics Research 14(1), pages 19-36, as follows: the arrangement of the machine parts at the time of collision (position of the axes), the dynamic mass properties, the kinematic structure of the robotic device (geometric arrangement of the axes, etc.), the collision direction, and the collision point. The dynamic mass properties of the robotic device may be calculated with or without the inertia of the drive. This gives either the effective mass of the robotic device without considering the drive inertia, or the effective mass of the robotic device with the drive inertia taken into account, and the effective drive mass is calculated from the difference between the two effective masses. When calculating with the drive inertia in mind, the transmission ratios of the individual drives of the robotic device must be taken into account. Khatib's method is particularly suitable for calculating the serial arrangement of machine parts that move relative to each other, for example, different elements of a robot chain in an articulated robot. However, the method can also be used for parallel kinematic chains and mixed configurations.

[0042] The effective stiffness (n) of the robotic device may be calculated according to J.-K. Salisbury, "Active stiffness control of a manipulator in cartesian coordinates," published in Proceedings of the Conference on Decision and Control including the Symposium on Adaptive Processes, pages 95-100, 1980, and A. Albu-Schaffer, M. Fischer, G. Schreiber, F. Schoeppe, and G. Hirzinger, "Soft robotics: what cartesian stiffness can obtain with passively compliant, uncoupled joints?", published in Proceedings of the 2004 IEEE / RSJ International Conference on Intelligent Robots and Systems (IROS), Sendai, Japan, February 2004, pp. 3295-3301, Vol. 4, depending on the arrangement of the mechanical parts at the time of collision, the stiffness of the individual drive trains, the kinematic structure of the robotic device, the collision direction, and the collision point. In these articles, according to an equivalent model, the effective driving mass is elastically coupled to the effective mass of the shaft, as for example shown accordingly in Figure 3. The stiffness between these masses then results in the effective stiffness.

[0043] To determine the forces in the case of a free impact (impact without clamping), the effective mass of the human according to (n) is required. The effective mass of the human is calculated here using a dynamic mass model of the human or is selected from a tabular list, for example according to ISO / TS15066. The dynamic mass model usually considers the body posture (i.e., the position of the individual joints), dynamic mass properties, such as dynamic mass properties that depend on gender, height, and weight, the impact direction, and the impact point. The tabular list usually includes the affected body segments and body postures. In the case of a clamping impact, the part of the human body (body part) cannot avoid the applied force, i.e. cannot retreat. Therefore, in the equivalent model, the body part must be considered as being rigidly trapped.

[0044] Using certain variables as parameters, a simulation model of a free impact model or a clamping impact model may be built, for example as shown in FIG. 3. Thus, the free impact model may include or be a three-mass oscillator model, and / or the clamping impact model may include or be a two-mass oscillator model. These models are suitable to draw conclusions about the force curves at the contact point based on the impact speed. The transmitted energy and the maximum force may be determined from the force curves and then used to draw conclusions about the expected injury via the resulting stiffness characteristic curves, and accordingly a safe speed may be predefined as the maximum allowed speed.

[0045] The calculation of the safe speed as the maximum allowable speed is analytically possible here if the resulting stiffness characteristic curve at the contact point is linear, and only numerically possible if the resulting stiffness characteristic curve at the contact point is nonlinear. In order to accelerate the method here, it may be possible to linearize the resulting stiffness characteristic curve at the intersection of the resulting stiffness characteristic curve and the limiting value of the biomechanical force. In this case, for example, an initial estimate of the maximum allowable speed may be derived based on an analytical solution and then used as an initial value for the numerical solution.

[0046] It should be noted here that in case of a temporary, i.e. relatively fast, impact, the drive controller of the robotic device may be ignored since the impact duration is usually too short for the drive controller to intervene. Therefore, the set and monitored reaction forces of the machine usually do not influence the impact. In case of an impact, biomechanical limits may be respected which are usually smaller than the set and monitored reaction forces of the robotic device.

[0047] Another aspect also relates to a control unit for predefining a maximum allowable speed of a robotic device. Such a control unit comprises a detection unit for detecting a contact point between a human operator and the robotic device, for detecting a shape of the robotic device at the contact point, and for detecting a spatial boundary condition of the collision, in respect of a collision between the human operator and the robotic device. Furthermore, the control unit comprises a computing unit for determining whether the collision is a non-clamping collision or a clamping collision, taking into account the physical boundary conditions, and for calculating a maximum allowable speed of the robotic device at the contact point using a free impact model if the collision is a non-clamping collision and using a clamping impact model or a quasi-static clamping model if the collision is a clamping collision. These models are different models in each case. The computing unit is also configured to output a control signal depending on the calculated maximum allowable speed of the robotic device. A further aspect relates to a robotic device comprising such a control unit.

[0048] Advantageous and advantageous embodiments of the control unit and of the robotic device comprising the control unit correspond to the advantageous and advantageous embodiments of the methods described herein.

[0049] In addition to the features and combinations of features mentioned above, both in the description and in the introduction, the features and combinations of features described below in the description of the figures and / or the features and combinations of features shown alone in the figures may be used in the combinations shown in each case, but also in other combinations, without departing from the scope of the present invention. Thus, embodiments that are not explicitly shown and described in the figures, but which are evident from the described embodiments and which may be produced by separate combinations of features, should also be considered to be included and disclosed by the present invention. Thus, embodiments and combinations of features that do not include all the features of the originally formed independent claims should also be considered to be disclosed. Moreover, embodiments and combinations of features that go beyond or deviate from the scope of the combinations of features presented in the reference to the claims should be considered to be disclosed, particularly by the embodiments presented above.

[0050] The subject matter according to the invention is explained in more detail on the basis of the following figures and schematic diagrams, which do not wish to be limited to the specific embodiments shown therein. [Brief description of the drawings]

[0051] [Figure 1] FIG. 1 is a diagram of an exemplary robotic device including an exemplary embodiment of a control unit for predefining a maximum allowable speed of the robotic device. [Diagram 2] FIG. 13 is a diagram of an exemplary resulting stiffness characteristic curve at a contact point. [Diagram 3] 1A-1C are diagrams of alternative models for an exemplary free impact model and an exemplary clamping impact model. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0052] In the different drawings, similar or functionally similar elements are given similar reference numbers.

[0053] Fig. 1 shows a robotic device 1 including a control unit 2 for predefining the maximum allowed speed of the robotic device (1). The control unit 2 includes a detection unit 3 for detecting the contact points between a human operator and the robotic device 1, the shape of the robotic device at the contact points and the spatial boundary conditions of the collision, in this case for a collision between the human operator and the robotic device 1. In the shown example, the contact points a, the spatial conditions b and the shape c are predefined by a user 5 of the control unit 2.

[0054] The control unit 2 also includes a computing unit 4 for determining whether the collision is a non-clamping collision or a clamping collision taking into account the spatial boundary conditions b of the collision and for calculating the maximum allowable velocity of the robotic device 1 at the contact point a using a free impact model if the collision is a non-clamping collision and using a clamping impact model or a quasi-static clamping model if the collision is a clamping collision.

[0055] These models are different models in each case. The computing unit 4 is here also configured to output a signal g that depends on the calculated maximum allowed speed of the robotic device 1.

[0056] In the illustrated example, the control unit 2 retrieves the joint configuration d from the robotic device 1 with respect to the time of impact and tool data e representing the tool of the robotic device 1. In addition, the torque or force thresholds f are also retrieved from the robotic device. In the illustrated example, the computing unit 4 retrieves from the machine database 5 the predefined kinematic structure of the robotic device 1 as well as other data of the robotic device, such as mass, inertia, center of gravity of the joints and motors of the robotic device in the illustrated example. Here, torque data is also retrieved. The stored models, i.e. the free impact model, the clamping impact model and the quasi-static clamping model, are retrieved accordingly from the model database 6 and the associated biomechanical database 7, together with the associated values ​​such as limit values, stiffness characteristic curves and masses of the body parts affected by the impact.

[0057] Figure 2 shows an example of the resulting stiffness characteristic curve h as a function of the force F on the formation D. The allowable penetration depth x2 corresponds to the deformation D at the intersection of the allowable biomechanical contact force y2 and the resulting stiffness characteristic curve h. The reaction distance x1 of the robotic device is obtained from the set reaction force y1 of the robotic device and the resulting stiffness characteristic curve h. The allowable braking distance of the robotic device is then obtained from the allowable penetration depth x2 minus the reaction distance x1.

[0058] Figure 3 shows an alternative model for the free impact model and the clamping impact model. In both cases, the robot device 1 moves at the impact point with a collision velocity v C In this specification, the effective driving mass m D is the effective stiffness of the drive system c T Through the effective mass m of the shaft L and then move with the effective stiffness c M The contact force F(t) is transmitted through the effective mass m D , m L But at different distances x D , x L You may only move

[0059] In the three-mass oscillator model, as shown in the upper right corner of Fig. 3, the finite effective mass m H But human stiffness c H Thus, in the example shown, the free impact model is D , m L , and m H In the case of the pinching impact, it is assumed that the operator 5 does not move, as shown in the bottom right of Figure 3. Therefore, the human stiffness c H absorbs the entire contact force F(t). In this method, two masses m D , m L Since only is considered, the impact model to be tightened is a two-mass oscillator model.

Claims

1. 1. A method for predefining a maximum allowable speed of a robotic device, the method comprising: - a method step of predefining a contact point between a human operator and the robotic device, a shape of the robotic device at said contact point and spatial boundary conditions of said collision, for a collision between said human operator and said robotic device; a method step of determining, using a computing unit, whether said collision is a non-pinching collision or a pinching collision, taking into account said spatial boundary conditions; calculating, by said computing unit, the maximum allowable velocity of the robotic device at the contact point using a free impact model if the collision is a non-clamping collision, using a clamping impact model if the collision is a clamping collision or using a quasi-static clamping model, said models being different in each case; a method step of outputting, by said computing unit, a signal dependent on said calculated maximum allowed speed of said robotic device; A method comprising:

2. 2. The method according to claim 1, characterized in that the output signal represents a position-dependent speed specification along a predefined machine path for the robotic device, the speed specification being generated by modifying the magnitude of an originally predefined speed specification for the predefined machine path, in particular based on the calculated maximum speed, preferably taking into account a predefined process speed of the robotic device in one or more subsections of the machine path, the modification of magnitude may comprise a uniform modification of magnitude or a modification of magnitude locally adapted to the predefined machine path.

3. 2. The method of claim 1, wherein said maximum allowable speed is calculated in real time, and said output signal represents an instantaneous maximum allowable speed of said robotic device.

4. 3. The method of claim 1 or 2, characterized in that if the collision is a clamping collision, the maximum allowable velocity is calculated using the clamping impact mode and the quasi-static clamping model, and the lower of the calculated maximum allowable velocities is selected as the maximum allowable velocity for the output.

5. In the quasi-static tightening model, the maximum allowable speed is (a) a predefined kinematic structure of the robotic device; (b) a joint configuration, including in particular the positions of one or more axes of the robotic device and the velocities assigned to each of said axes, at the time of said collision; (c) a stiffness characteristic curve of the body part at the contact point of the impact; (d) a stiffness characteristic curve of a point of the machine at the contact point of the impact; (e) a resultant stiffness characteristic curve calculated using the stiffness characteristic curve of the body part and the stiffness characteristic curve of the machine point; (f) a predefined maximum allowed deformation, in particular using a force threshold and / or an energy threshold and / or a deformation threshold; (g) an allowable penetration depth calculated using the stiffness characteristic curve of the machine point and the resulting stiffness characteristic curve; (h) a reaction force of the robotic device; (i) a reaction distance of the robotic device calculated using the stiffness characteristic curve of the machine point and the allowable penetration depth; (j) an allowable braking distance of the robotic device calculated using the predefined maximum allowable deformation and the reaction force; and (k) the actual braking distance of the robotic device.

3. The method according to claim 1, wherein the calculation is based on:

6. 6. The method according to claim 5, characterized in that the maximum permissible speed of the robotic device is determined iteratively, in particular by using a bisection method, until the actual braking distance coincides with the permissible braking distance.

7. 6. The method of claim 5, wherein if the crash involves contact points at different body parts, the output signal depends on the maximum allowable speed that corresponds to the shortest actual braking distance.

8. 6. The method of claim 5, wherein if the collision involves contact points at different machine points, the maximum allowable speeds of all contact points are calculated and the output signal depends on the lowest of the maximum allowable speeds.

9. In the case of the free impact model and the clamping impact model, (a) the predefined kinematic structure of the robotic device; (b) a joint configuration, including in particular the positions of one or more axes of the robotic device and the velocities assigned to each of said axes, at the time of said collision; (c) a stiffness characteristic curve of the body part at the contact point of the impact; (d) a stiffness characteristic curve of a point of the machine at the contact point of the impact; (e) a resultant stiffness characteristic curve calculated using the stiffness characteristic curve of the body part and the stiffness characteristic curve of the machine point; (f) a predefined maximum allowable deformation using a force threshold and / or an energy threshold; (l) the effective mass of the machine at the contact point of the impact; (m) the effective stiffness of the robotic device, and, for the free impact model only, (n) the effective mass of the body part at the contact point of the impact 3. The method according to claim 1, wherein the maximum permitted speed is calculated in each case on the basis of:

10. 10. The method of claim 9, wherein the free impact model comprises or is a three-mass oscillator model and / or the clamping impact model comprises or is a two-mass oscillator model.

11. A control unit for predefining a maximum allowable speed of a robotic device, comprising: a detection unit for detecting a contact point between a human operator and a robotic device, a shape of the robotic device at said contact point, and spatial boundary conditions of said collision, for a collision between the human operator and the robotic device; a computing unit for determining, taking into account the spatial boundary conditions, whether the collision is a non-pinching collision or a pinching collision and for calculating the maximum allowable velocity of the robotic device at the contact point using a free impact model if the collision is a non-pinching collision and using a pinching impact model or a quasi-static pinching model if the collision is a pinching collision, wherein the models are different in each case, and a computing unit for outputting a signal dependent on the calculated maximum allowable velocity of the robotic device.

12. A robotic device comprising a control unit according to claim 11.

Images